Monetary policy and household consumption
Paul Schmidt
2025-03-17
1 Introduction
In recent years, both private consumer spending and the European
Central Bank’s (ECB) monetary policy have undergone significant changes.
Private consumption initially declined in 2020 due to the COVID-19
pandemic but subsequently increased steadily. However, between 2022 and
2023, consumer spending remained relatively stable. At the same time,
the household savings rate decreased continuously, falling from 15.3% in
2020 to 10.4% in 2023 (cf. Destatis, 2024). In response to financial
stress during the pandemic and the subsequent inflationary period, the
ECB implemented various monetary pol icy measures. During this period,
monetary policy underwent notable shifts. In July 2022, the ECB raised
interest rates for the main refinancing operations (MROs), the marginal
lending facility, and the deposit facility for the first time since
2011. Within a year, these interest rates reached levels comparable to
those ob served during the euro crisis in 2008 or shortly after the
euro’s introduction in 2001 (cf. ECB, 2024). Additionally, the volume of
open market operations (OMOs) increased significantly in 2020, primarily
due to the expansion of the asset purchase programme (APP) and the
introduction of the pandemic emer gency purchase programme (PEPP). The
ECB also adjusted its set of monetary policy instruments. In April 2020,
the pandemic emergency longer-term refinancing operations (PELTROs) were
introduced to stabilize liquidity conditions in the euro area and
support money market dynamics at the onset of the COVID-19 crisis
(cf. ECB, 2024). Alongside PELTROs, the ECB launched the PEPP as a
response to the pan demic. Furthermore, net purchases under the APP,
which had been in place since 2014, were discontinued in July 2022
(cf. ECB, 2024). Given these developments, this thesis examines the
impact of the ECB’s mon etary policy from 2020 to 2023 on household
consumption. The analysis fo cuses on four key transmission channels:
the real interest rate channel, the wealth channel, the household
liquidity channel, and the credit channel. Un derstanding how these
channels influenced consumer behavior provides 7 valuable insights into
the effectiveness of monetary policy in times of economic uncertainty.
Figure 1: Monthly change of the ECBs OMOs 01.04.2015 –
01.11.2024. Source: ECB2, 2024
Notes: The APP started 2014 but the
first purchases were in March 2015, the chat presents the change of
monthly net purchases of Euro-system holdings under the APP of the
ECB
2 Literature review
The monetary policies of the ECB are directly interrelated with price
changes. In the case of the ECB, the mandate is price stability, and
central banks such as the Federal Reserve (FED), the Reserve Bank of
India (RBI) and the Peo ple’s Bank of China (PBC) also have price
stability as one of their main ob jectives (cf. FED, 2020; PBOC, 2018,
RBI, 2021).
The Phillips curve was used by Alban William Housego Phillips in 1958 to
describe the negative non-linear relationship between the unemployment
rate and monetary wage increases. The modified Phillips according to
Samelosn and Solow was used to describe the relationship between the
inflation rate and the unemployment rate. (cf. Wohltmann, 2024) In
recent decades, the Philips curve has been further developed, including
with the Neo-Keynesian approach, where the model is extended to include
information asymmetries and labor market dynamics. Another extension is
the agent-based model (AB model), which assumes additional budget
differences among households. 8 Based on the AB model, Lilian Rolim,
Laura Carvalho and Dany Lang show in the paper “Monetary policy rules
and the inequality-augmented Phillips curve” (2024) that the flattening
of the Phillips curve could be related to the flattening of purchasing
power and the increase in income inequality. On the other hand, they
show that monetray policy that prioritizes lower inflation rates leads
to increased unemployment and income inequality. (Rolim et.al., 2024)
Accordingly, income inequality is an important indicator of the impact
of monetary fluctuations on the economy.
The AS-AD model is an important link in deriving the interaction between
the economy and the monetary policy of the central banks. With this
model, which also implies the IS-LM model, it is possible to
theoretically derive the effects of expansionary and restrictive
monetary policy. However, the ECB’s monetary policy does not have a
effect on the economy, only through certain transmission channels. An
example is the direct influence on the expectations of households or
household consumption. Bernd Hayo describes this in the study “Does the
ECB’s monetary policy affect personal finances and eco nomic inequality?
A household perspective from Germany” (2023) using a data set from 2018
and comes to the conclusion that the perception of the ECB’s monetary
policy has no direct significant influence on household in come or
household consumption. It is established to distinguish between dif
ferent monetary policy transmission channels. In “Monetary Development
and Transmission in the Eurosystem“ Anton (2015) differences between 16
transmission channels of the ECB’s monetary policies. Anton divides the
MTCs (monetary transmission channels) into different ‘views’, the asset
view, traditional view, prospect view and the credit view. Mishkin
(2019) divides the MP transmission into eight MTCs. 9
Figure 2: Monetary policy
transmission channels according to Mishkin (2019). Source: Mishkin,
2019, p.681.
For this study the MTCs real interest rate channel,
financial wealth channel, the household liquidity channel and the
traditional credit channel are used. This MTCs are considered by Mishkin
(2019) and Anton (2015) as stand alone transmission channel. Mishkin
(2019) refers to the traditional credit channel of Anton (2015) as the
bank lending channel. Mishkin (1996) refers to real interest rate
channel as the MTC widely used in the traditional Keynesian literature.
The IS-LM model includes the mechanics of this transmission channel as a
reaction chain to expansive MP (monetary policy). This monetary
transmission channel is described by Mishkin (1996) as follows:
M ➔
ir ➔ I ➔ Y .1
In this channel of the real interest rate, Mishkin
(1996) mentions, among other things, sticky prices as a causal factor,
as well as consumption expenditure 1M : expansionary MP, ir : decrease
in real interest rates, I : increase in investment spending, Y :
increase in aggregate demand and output. 10 and investment. However,
consumption expenditure is only added to the real interest rate channel
by Anton (2015). Anton (2015) rates this channel as one of the most
important MTCs in the long term and also points out that the pass
through rate is important for the efficiency of this channel. Gomes and
Seoane (2024) come to a similar conclusion in their study “Made in
Europe: Mone tary-Fiscal policy mix with financial frictions” on the
differences and their effects of monetary policy during the euro crisis.
However, the transmission channel is also questioned, as Rubert and
Sustel (2019) have presented the MTC observationally, but not
structurally, in the paper “On the mechanics of New-Keynesian models”,
since the real interest rate represents the feasibility to keep
consumption smooth and therefore contractionary monetary policy is only
consistent with real interest rate changes, but is not causal. They
argue that, according to their model, the expectation of monetary policy
shocks is more relevant for the impact of MP on household consumption,
referring to the Lucas Critique. In contrast, Kawamato et al. (2023)
were able to show in their paper “Estimating the macroeconomic effects
of Japan’s expansionary monetary policy under Quantitative and
Qualitative Monetary Easing during 2013-2020” using a counterfactual
analysis that the real interest rate channel was significant for the
change in real GDP during the transfer of open market operations of the
Bank of Japan. The financial wealth channel describes that an
expansionary monetary policy can lead to an increase in the stock
prices, this can lead to an increase in con sumption and this to an
increase in the GDP. Anton (2015) differs consump tion in the
consumption of durable goods and residential investments. This monetary
transmission channel was strongly influenced by Modiglianis life cycle
model as Mishkin (1996) describes. Albacete and Lindner (2017) were able
to find a limited long-term correlation between for the wealth channel
in Austria for the period 2010 to 2014. They also conclude that the
effect of the wealth channel is heterogeneous among households. The
marginal propensity to consume from wealth is responsible for the
limited significance of the wealth channel due to the fact that the
consumption function is concave, and wealthier households have a lower
propensity to consume in the context of unequal wealth distribution. In
the study “Monetary policy, asset prices and 11 consumption in China”
(2012) by Koivu, the reaction of household consump tion to monetary
policy was examined for the years 1998 to 2008 in China. Koivu (2012)
concludes that the influence of the wealth channel on consump tion is
relatively small here, as the changes in wealth from the household per
spective are relatively small. Overall, the wealth channel appears to be
limited here and there are indications that it could have a weak impact
on residential prices in China in the long term. The heterogeneity of
the households is also seen as a relevant factor for monetary policy by
other studies. For example, in the study “Does wealth inequality affect
the transmission of monetary pol icy?” (2023) showed that in the US and
UK in the time period from 1969 to 2007 and in the euro area from 1999
to 2020 the effectiveness of monetary policy on real variables such as
GDP or unemployment increased with rising income and wealth inequality.
In their study “The impact of monetary policy shocks on net worth and
consumption across races in the United States” from 2024, Albert and
Gómez-Fernández also found a different effect of expan sionary monetary
policy and income groups for the years 1990 to 2020. In the case of an
expansionary monetary policy, the volatility of household con sumption
increases with household wealth. Furthermore, Alp and Seven (2019) were
also able to prove the significance of the wealth channel for Tur key
from 1998Q1 to 2016Q2. The interest rate driven shocks are particularly
significant for the wealth channel. In addition, also relevant to the
liquidity channel, asset prices are influenced by an interest rate
shock. The liquidity channel, or more precisely according to Anton
(2015) the house hold liquidity channel, works via household assets,
which then have a nega tive impact on the financial distress of
households, leading to a better expec tation of a more consumer-friendly
propensity to consume. This leads to an increase in household
consumption. In their paper “Quantitative easing with heterogeneous
agents”, Cui and Sterk (2021) analyzed the US economy from 2008 to 2016.
They found that during the recession period, quantitative eas ing (QE)
MPs had an impact on household consumption through the liquidity
channel. Herrenbrueck (2019) also associates the liquidity channel with
cen tral bank open-market operations (OMOs). From his model with
heterogene ous households, he concludes that OMOs can indirectly cause
crowding-out 12 effects in more liquid assets. Furthermore, if there is
a strong incentive to buy assets, money rotation can be reduced, which
can lead to disinflation. The traditional credit channel contains a
positive relation between expansion ary monetary policy and the volume
of loans, due to an increase in the bank’s reserves and banks deposits.
This leads to an increase in investments and in consumption. Sapriza and
Temesvary (2024) analyzed the influence of the credit channel for the US
economy from 1986 to 2019. they found an in creased significance of the
credit channel for the transmission of monetary policy during periods of
lower growth. Berqiraj et al. (2025) come to a similar conclusion for
the US economy from 1979 to 2015. According to them, the credit channel
is particularly powerful contractionary MP shocks. Evgenidis and
Salachas (2019) come to a similar conclusion for the euro area for the
time period 2003 to 2017. During periods of financial stress the demand
for loans increases and, together with expansionary unconventional MPs
the credit channel becomes more effective.
The following MTCs are assumed, as per Anton (2015), and are to be used
for further analysis:
Real interest rate channel:
M → r → [I + C durable, housing ] → Y ,
Financial wealth
channel:
M → Pstock, Houses, Land → wealth → ( Cdurable goods +
Iresidential ) → Y ,
Household liquidity channel:
M →
household’s financial assets → financial distress → C → Y ,
Credit
channel:
M → reserves → bank deposits → volume of bank loans → I → C
→ Y .2 2
An increase in M describes expansionary monetary policy. C
describes the consumption, I investments, P the prices and Y the GDP. 13
In the past, other channels have also been examined specifically for
their ef fect on household consumption, such as the cash flow channel by
Hughson et al. (2016). However, the four mentioned so far appear to be
the most im portant. In addition to the extant literature on the
corresponding transmission chan nels, which mostly relates to the period
up to 2020, there are also some stud ies, albeit fewer, on the period up
to 2023. These latter studies address the more recent developments
accordingly. In „Who bears the costs of inflation? Euro area households
and the 2021–2023 shock” (2024) Pallotti et al. uncov ered sizable
average losses and a significant level of heterogeneity across countries
and, within countries, across age groups, but not across income groups.
These were the effects in Germany, France and Italy, only in Spain there
is a heterogeneity across income groups. The highest welfare loss were
in Germany and Italy, France and Spain had a welfare loss of around 3%.
Speaking of Spain Uxó et al. (2024) analyzed in “Prices, markups and
wages: inflation and its distributive consequences in Spain, 2021-2023”
the impact of recent inflation on the real wages and the income
distribution recent infla tion. They also analyzed the influence of
anti-inflationary economic policys, considering fiscal policy of the
Spanish government and monetary policy of ECB. Uxó, Febrero and Álvarez
considered the increasing of the interest rates as an inappropriate
policy, the argue that the Spanish inflation was not caused by excess
demand. They analyzed the effects of the fiscal policies and the
monetary policies had an opposite growth stimulus. Glocker and Wegmüller
share Uxó et al.’s criticism of the ECB’s monetary policy for 2021 and
2022 with their findings from the study. „Energy price surges and
inflation: Fiscal policy to the rescue?” of the november 2024. They
conclude that a taylored fiscal policy is superior to a demand-side
policy that faces a supply-side shock. With the aim of price stability,
they see fiscal policy measures as more efficient than MPs. The
following research question arises from the literature and developments
in recent years: 14 What was the influence of the ECB’s monetary policy
on household con sumption in the Eurozone in the years 2020–2023? This
research question is formulated more generally; there are several sub
questions in this research question. The first question is how relevant
the real interest rate channel is. It can be assumed that it is relevant
overall, as real interest rates are relatively oriented towards the MRO.
However, this interest rate channel has usually been the most
significant in the long-run. Further more, the question arises as to how
relevant the wealth channel is, as this is very closely linked to the
wealth distribution of households. With the liquidity channel, the
question arises as to how relevant this is, as it has less evidence in
comparison. If the Lucas Critique implication of Rubert and Sustel
(2019) is taken into account, according to which the expectation of an
MP shock is relevant for MP transmission, this transmission channel
household financial distress can be significant. It is also interesting
to see how credit channel in teracts, as the literature of the last 10
years suggests that it is particularly sig nificant for monetary
tightening and financial stress. Furthermore, it is ques tionable
whether transmission channels have worked in the same way in every EU
country, as each EU country can pursue its own fiscal policy. If we add
to this the results of Uxó et al. (2024) and the results of Duarte and
Pereira (2022). In their paper “The effect of monetary policy on
household consump tion expenditures in Portugal: A decomposition of the
transmission channel”, Duarte and Pereira (2022) were able to show that
the Portuguese economy, like the Spanish economy, reacts more strongly
to monetary shocks than France or Germany due to a differentiated
household consumption structure.
3 Methology and data
The data used in this study is sourced from Eurostat and the ECB
database for the period from 01.01.2020 to 31.12.2024. The primary focus
is on changes in the ECB’s monetary policies. An overview of all
variables, includ ing their sources and frequencies, is provided in the
appendix A.1: along with a detailed description of some variables, under
appendix A.2. 15 For the following methods, I utilize data from the ECB,
Eurostat, and the IMF database. The period under examination extends
from the beginning of 2020 to the end of 2024, with most data sampled in
monthly intervals. The analysis focuses on the euro area, as the ECB’s
monetary policies are designed for the entire region, making it the
relevant reference framework. However, the com position of the eurozone
changes during this period, as Croatia became a member of the eurozone
and the Schengen area in January 2023. Most of the used data is monthly,
but also some data time lines are only available in the quarterly data,
like the Gini or the equity data. Because the transmission chan nels I
analyze here are based on linear connections, it is still possible to
extract a positive or a negative relationship. But the time dependent
analysis is there fore to some degree biased. The data for countries
Germany, Spain and France is based on the same data sources as the
aggregated euro area data. This is necessary to create comparability
between the different results.
The methodological approach is structured as follows. First, I analyze
the im pact of monetary policy on household consumption through the MRO
and OMO channels. Subsequently, household consumption is disaggregated
to ex amine distributional effects and consumer goods elasticities. I
then investi gated country-level effects to determine whether the
results vary across indi vidual eurozone nations. Additionally, fiscal
policies of member states are considered to validate the findings of Uxo
et al. (2024) and Pallotti et al. (2024) regarding the fiscal policy
effects in 2021 and 2022. I differ into four different types of monetary
policy transmission to household consumption. First, I use OLS
regressions to track the transmission channels and to track the entire
transmission channel via interaction terms. My meth odology in case of
the OLS regressions is based on the mediation analysis of a study by Du
et al. (2025) which they used to analyze the effects of monetary policy
on energy consumption. In their study, they used a two-stage approach.
First they analyzed the effect of MPs on the MTC - variables and the
effect of the MTC-variables on their variable of interest, in their case
energy pov erty. A similar methodology was also used by Tang and Yang
(2024) to ana lyze the effect of monetary policy uncertainty on
financial risk. I also use a two-stage approach for the first two
transmission channels. For the liquidity 16 channel and the credit
channel, I use a three-stage approach to take into ac count the
theoretical representation of the transmission channels. Second, I use
vector autoregressive (VAR) models to analyze the transmission channels.
These are often used in different variations in papers to analyze
monetary transmission channels. This is due to the simple implementation
of lags, the implementation of reciprocity of variables, as all
variables are con sidered endogenous, and the possibility to analyze
shocks using impulse re sponse functions (IRFs). I used the recently
published papers by Vale (2024), De Simone (2024) and Mundra and Bicchal
(2023) as a reference. The regres sion equation for the VAR model is as
follows: Xt = A (L) Xt + ɛt. (1)
X represents the endogenous variables. An important prerequisite for a
VAR model is stationarity. I check this stationarity using the augmented
dickey fuller (ADF) test. If the time series of the data frame are not
stationary, I tests whether they can differentiate them to establish
stationarity, or whether they are also cointegrated. This test is
performed using the Johansen test. If the time series are cointegrated,
then a VECM model is used. This is similar to the approach of Sun et
al. (2010), who analyzed the credit, interest rate and asset price
channels for China from 1996 to 2006.
The VECM model can be described as follows: ∆𝑥𝑡 = П𝑥𝑡−1 +∑ Г𝑙 𝑝−1 𝑙=1
𝛥𝑥𝑡−𝑙 + 𝐶𝑑𝑡 + ɛ𝑡. (2) Δx describes the difference variables and Г
describes the coefficient matrix. The rank of П describes r, so the
number of cointegration relationships of variables. The VECM can now be
estimated. The short-term and long-term dynamics of the variables can be
determined. For further analysis, I created IRFs for the relevant
interactions of the MTCs. To better compare the results of the four
MTCs, I use the Granger test.
17
4 Monetary Transmission channels
In order to examine the effects of monetary policy in more detail, I will now focus on the transmission channel of monetary policy, which is subdivided by Anton (2015), using the real interest rate, financial wealth, the household liquidity and the credit channel.
4.1 Real interest rate channel
Before we start with the implementation of further different channels
we an alyse the direct effects of the ECB monetary policy on Household
consump tion. The connection between monetary policy and consumption I
am looking at, is the real interest rate channel which is based on the
mechanics of the IS LM-model and AS-AD-model. Also, Anton (2015) refers
to this channel as his first transmission channel. Anton (2015)
considers this transmission chan nel as one of the strongest
transmission channels, because of the long run impact on the real GDP.
Therefore, the basic assumption of the relation is that expansive
monetary policy has a negative influence on the real interest rates, the
real interest rates have a positive impact on investments and
consumption of durable goods.
Therefore, I analyze first the direct effect of the main refinancing
operation (MRO) policy and the regular and non-regular open market
operations (OMO)3 on the real interest rates. The real interest rates
can be measured by the nominal interest rates and the inflation: Real
interest rate = nominal interest rate – rate of inflation. For the
calculation of the real interest rates (r), I use the interest rate for
house holds for borrowing money for a house purchase and the interest
rate for the interest rate for loans with agreed maturities for
households. I am not taking the ECB data for the interest rates for
consumption into account because this includes also interest rates for
loans for the consumption of non-durable goods. I am also not taking
interest rates for cooperations into account 3In the period from 2020 to
2023 non-regular open market operations, APP, PELTROs and PEPP had high
volumes, see Figure 1: Monthly change of the ECBs OMOs 01.04.2015 –
01.11.2024. 18 because here is the focus on the households. To determine
the inflation, I use the harmonized index of consumer prices (HICP) as
an indicator. Also, be cause it is close to the actual price dynamics of
the households and other stud ies, like Pallotti et al (2024), who
estimate also the price changes in euro area with the HICP. Pallotti et
al (2024) use for their estimation of household con sumption the data of
the household budget survey (HBS). Unfortunately, I cannot use this
data, because the dataset has a five-year frequency (cf. Euro stat,
2025), but here it is the monthly frequency we need. The data I use is
quarterly data on the aggregate of final consumption from Eurostat. Here
I estimate the monthly consumption via the quarter-to-quarter trend. I
also dif ferentiate between final consumption of all goods and final
consumption of durable goods from 2022 Q1 to 2024Q3. For the entire time
period, I differ entiate between aggregated consumption and the
consumption of gas and electricity.
In Figure 3 and Figure 4 the red graph describes the real interest
rate for the households for loans for buying houses and the orange graph
describes the real interest rate for loans with fixed maturities. The
orange and red graph describe the real interest rate transmission
channel and are therefore visual ized in both figures. On the left, next
monetary policies during the period of 2020 to 2024 and on the right
next to different consumption quantities. Inter actions can be guessed
at in the two figures, for example that in Figure 3 the real interest
rates at the end of 2022 react positively to the increase in the MRO.
However, interpretations from the visualizations remain only approxi
mations, especially not visible, for example aggregate consumption, it
is in 19 the nature of things that this is not as volatile as the other
consumption sub categories and since in Figure 4 only shows marginal
consumption data, it is difficult to gain meaningful insights here. The
data visualized in Figure 3 can be described by the following regression
equation: rt = α + β1 × MROt + β2 × OMOt + ɛt. (3) I estimate the
regression two times, to consider the two different variables for the
real interest rate. Once for the real interest rate for house loans for
house holds and once for the real interest rate.
The second step is to examine the effects of changes in real interest
rates on consumption. In the case of the real interest rate, a
distinction is again made between the real interest rate for home loans
(r14) and the real interest rate for loans with an agreed maturity (r2).
The regression looks like this: consumptiont = α + β1 × r1t + β2 × r2t +
β3 × (MRO × r1t) + β4 × (MRO × r2t) + β5 × (OMO × r1t) + β6 × (OMO ×
r2t) + β7 × kt + ɛt. (4) In addition to the influences of r1 and r2
directly, I refer to the basic assump tion of the transmission channel
that the MPs become effective through the transmission channels. To
include this, I implement interaction terms of r1 and r2 with MRO and
OMO in (4). Thus, the influence of r1 and r2 can be captured with
changing MPs. The variable consumption is estimated several times, once
with the consumption aggregate, the consumption of durable goods and the
consumption of gas and electricity respectively. The variable k stands
for the control variables, in (4) this is unemployment. The results for
the first regression model (3) are that both real interest varia bles
are significant to 0.01 level to the MRO-variable and to the 0.05 level
significant to the OMO-variable.5 The results for the second regression
model (2) differ considering what endogenous consumption variable I
used. For 4In the following, the variables are always written in
italics, this is relevant when differenti ating between the variables
MRO and OMO and simple abbreviations MRO and OMO. 5The complete results
of the regressions for (3) can be seen in the appendix under B.1 and
B.2. 20 second regression models that follow the structure of (4) for
the durable goods consumption the real interest rate with fixed
maturities is significant. In terms of regressions with the consumption
of electricity and gas the MRO and the interactions terms of the MRO
with the real interest rates were significant. The coefficient
differences between the real interest rates in both cases. The real
interest rate for house loans is in both cases positive, which indicates
a positive influence of an increase in the real interest rates of house
loans to positive change in terms of the effectiveness of an MRO-change.
The real interest rate of the agreed maturity has in both cases a
negative coefficient, which means that effectiveness of the MRO
decreases with an increase of the real interest rate of the loans for
households with agreed maturities. Although these interaction variables
are significant in the other two regressions, there is also a difference
in the coefficients. The interaction terms with the OMO variable are not
significant in any of the regressions; if you look at the coef ficients
of these terms, you will notice that the signs of the coefficients are
reversed compared to the MRO interaction terms. In the regression with
final consumption as the endogenous variable, no variable other than
unemploy ment is significant. However, since the MPs do not act directly
via transmis sion channels and consumption has a temporal inertia, it
makes sense to in clude temporal lags.6 Furthermore, it makes sense to
consider other variables and then possible endogeneity problems. I use
an autoregressive (AR) model for this purpose. To get more information
about the dependencies, trends, number of lags, sea sonality and further
indicators for different models I analyze at first the auto correlation
functions (ACF).
The time series for final consumption and MRO shows a slow decline,
which could indicate a trend in the data. It is also obvious that there
is seasonality in gas and electricity consumption. This is indicated by
the ACF, can be guessed from the graph and would also be consistent in
terms of content. Since I plan to use a VAR model similar to Vale
(2024), the data must be examined for stationarity. This is particularly
important because there is a trend in the final consumption variable and
the MRO. I use the augmented dickey fuller test to check the
stationarity. The p-values for all relevant variables from the 22
augmented Dickey-Fuller test are shown below. If the p-value p <
0.05, the process is stationary; if the p-value p ≥ 0.05, the time
series is non-stationary.
The time series of the variables MRO, OMO, r1, r2, final consumption
and durable goods are stationary. There are some options to estimate an
auto regressive model for a non-stationary time series.7 It is possible
to differenti ate the time series or to estimate another model like a
vector error corrected model (VECM). To find out whether it is
sufficient to differentiate the time series, a cointegration test must
be carried out, for this I use the Johansen test. Since there is more
than one cointegration relationship at a critical value of 0.01, 0.05
and 0.1, a VECM model is used instead of differentiation.8
I estimate the VECM with two control variables, in addition to
unemployment I also imply the Gini coefficient. I imply the Gini
coefficients, since according to Matusche and Wacks (2023) the degree of
income inequality has an impact on the effectiveness of monetary policy
transmission. The VECM model can be used to approximate the short-term
and long-term dynamics of the time series. For long-term dynamics I
evaluate the beta statistics of the VECM model.9 Thus, the variable
final consumption has a long-term positive effect on unemployment and
the consumption of electricity and a long-term nega tive effect on Gini,
on r2, on MRO and on the consumption of gas. If we now filter the
long-term dynamics of the variables that are relevant for the real
interest channel, interesting correlations emerge. If the MRO changes by
one unit c.b., then OMO, r1 and r2 behave exogenously in the long run
with re spect to a cointegration relationship. The r2 reacts negatively
to OMO in the 7The detailed results of the ADF tests can be found in the
appendix under B.8. 8Further results of the Johanson tests can be found
in the appendix under B.9. The Johanson test indicates a VECM model with
r = 7. 9The beta statistic describes how the variables are related to
each other in the long term, the complete matrix of VECM model beta
statistics is available under appendix B.10. 23 long term. Final
consumption and electricity consumption react positively to r1 and r2 in
the long term, while gas consumption reacts negatively. Short term
dynamics can be analyzed in a similar way. The MRO has a short-term
negative impact on all variables, the negative impact on real interest
rates is relatively similar. The OMO also has a negative influence on
real interest rates in the short term, albeit a minor one. The real
interest rates have a different influence on final consumption. The
influence of r1 is positive and of r2 neg ative. Both variables have a
negative influence on the consumption of elec tricity and only r2 on the
consumption of gas. So far I have not gone into the variable durable
goods any further, as this would have halved the total period of the
analysis. Now add this variable and check what interactions the variable
durable goods has with the other variables. Similar to final
consumption, du rable goods react positively to r1 and negatively to
r2.
To further analyze the dependencies of the variables, I look at the
impulse response functions. The first step is to check the transfer from
MPs to the real interest rate. Here, r1 and r2 should only differ
marginally, as these variables are intended to capture comparable
dynamics in this context.
The figures 7 and 8 clearly show that the real interest rate reacts
positively to a shock in the MRO with an interval of three periods,
i.e. three months. In the event of a shock in the OMO, a negative
reaction in the real interest rate can initially be seen, but this
becomes less pronounced and after 4 months this reaction in the real
interest rate is positive in each case. Now I analyze the second part of
the transmission channel for shock re sponses using impulse response
functions. For this part, the same data sample is used again, except for
the variable durable goods, as this covers the shorter time span from
2022 to 2024.10
Figure 14 illustrates the impulsive response function of MRO to final
con sumption. The marginal response of final consumption to a change in
MRO is 10 The most changes in the MRO are in the period from 2020 to
2024 at the end of 2022, so the assumption here is that the most
relevant information regarding shock dynamics can be obtained from the
data from the middle onwards. In addition, this analysis includes
durable goods despite the lower availability of data, as Anton (2015)
refers to durable goods in the real interest rate channel. 25 clearly
visible, with a response most likely to be seen 4 to 6 months after a
change in MRO. If we now take the reaction of final consumption to the
real interest rate for housing, Figure 11 here we see a strong positive
reaction in final consumption two to six months later. In figure 12
shows the negative reaction of final consumption to an increase in the
real interest rate for loans with agreed maturity, revealing a certain
ambivalence between consumption and interest rates. Different effects
could be causal here, the different use of these consumer loans is
probably causal. For example, substitution effects are conceivable here,
with households consuming instead of saving when real in terest rates
for house loans rise and an increase in term loans reduces con sumption,
as fewer households are prepared to take out a loan for their con
sumption. This substation effect of housing credit consumption and
non-hous ing consumption was recently shown by Horioka and Niimi (2020)
for the Japanese households for the time period 1970 to 2017. The
negative effect of an increase in real interest rates for credits with
agreed maturity and house hold consumption refers to budgeting of the
households. The IRFs for the consumption of electricity and gas are
available under appendix A.3. Finally, as part of the analysis of the
real interest rate transmission channels, I perform a breakpoint
analysis and on this basis I perform IRFs for the smaller data samples
in order to take into account possible differences in the correlations
during different time periods. The breakpoint analysis suggests at least
one breakpoint for the MRO variable in October 2022. The more
breakpoints selected, the more detailed the increase in MRO is tracked.
For example, July 2022, the first increase in the MRO, and March 2023,
when the MRO increased from 3% to 3.5%, are mentioned at two
breakpoints. For most of the variables, especially for MRO, OMO, r1 and
r2, the Residual Sum of Squares (RRS) and Bayesian Information Criterion
(BIC) show the best model with two breakpoints, so I will use the two
breakpoints for the estima tion of IRFs. The breakpoints of the impulse
variable are taken as the starting point for the IRFs.
26
I have tried to illustrate the second part of the transmission channel with du rable goods and final consumption. Overall, it is noticeable that most of the breakpoints of the transmission channel are around the MRO increase. The transmission channel is also easily recognizable because you can see periodic time differences between the effects of the MPs and r1 and r2 and the effect of r1 and r2 on consumption. It can be observed that r1 and r2 react in a manner consistent with the responses exhibited by MRO and OMO shocks. Furthermore, an analysis of the data indicates that the response of durable goods consumption to a shock in r1 or r2 appears to exceed the response of final consumption. This observation is consistent with the assumed channel connection and the hypothesis of the substitution effect.
4.2 Financial wealth channel
The second transmission channel I analyze after the real interest
rate channel is the financial wealth channel. The wealth channel, as
described by Mishkin (2022), or financial wealth channel, as described
by Anton (2015), refers to the increase in the prices of stocks and
equity that can then lead to an increase in wealth and then may this
lead to an increase in consumption of durable goods and residential
investments. Accordingly, it is important for this trans mission channel
to changes in stocks and equity and the only of the here an alyzed
transmission channels how accounts in the transmission of the MPs to the
household’s wealth. Vale (2024) used to capture the dynamics in 27
household wealth the nominal housing prices. Therefore, it is important
the changes households equity into account so it is possible to account
for the transmission channels functionality via the households
wealth.
First, we start with the OLS regressions. I analyze the transmission
channel with the help of the portfolio changes of households and the
equity and in vestment fund investments of households. The focus is here
on equity and investment variable and the portfolio variable can be seen
as an indicator for the household’s wealth. The portfolio variable is in
the balance sheet channel in the focus, but the balance sheet channel is
here not analyzed the focus of balance sheet channel on companies. I use
real GDP as a control variable to account for changes due to cyclical
changes. The regression equations are as follows: portfoliot = α + β1 ×
MROt + β2 × OMOt + β3 × realGDPt + ɛt, (5) Equityt = α + β1 × MROt + β2
× OMOt + β3 × realGDPt + ɛt. (6) Consumptiont = α + β1 × portfoliot + β2
× Equityt + β3 × (Equityt × MROt) + β4 × (Equityt × OMOt) + β5 ×
(portfoliot × MROt) + β6 × (portfoliot × OMOt) β3 × kt + ɛt. (7) The
monetary policy, MRO and OMO, are significant to the 0.01 significance
level in the relation to the Equity and investment fund changes of the
private households, as well as the portfolio (assets/liabilities)
changes of the private households. To measure consumption, I primarily
use final consumption and the consumption of durable goods in order to
be able to better compare the transmission channels later on and to be
able to measure the effect on house holds well through the quantity of
consumption and this transmission channel, similar to the real interest
rate channel, specifically affects the consumption of durable goods. If
the endogenous variable of the regression equation (7) is the aggregated
final consumption or the consumption of durable goods, the interaction
term of the portfolio variable and the equity variable with the MRO is
significant at the 0.05 precent significance level in both cases. This
indicates that the transmission channel for the period under review
primarily worked via the change in MRO and not directly via OMOs.
However, the two 28 endogenous variables differ in terms of the
coefficients of the interaction terms. Thus, the effect of the variable
MRO on the variable final consumption becomes weaker when the variable
equityandinvestment increases and stronger when the variable portfolio
increases. The variable durablegoods is more strongly influenced by a
positive change in the MRO of the equityandin vestment variable. The
effect of the portfolio variable decreases when the MRO variable
increases.11 These interaction effects can be seen in red in the
following figure 16.
Only the MRO interaction effects are shown here, as only these were signifi cant for the endogenous variables. First of all, it is clear that the transmission channel was probably less active via MRO changes before the interest rate hike in mid-2022. Furthermore, a temporal difference between the change in the MPs and the dynamics of the transmission channel can be inferred. This difference indicates a time delay within the transmission channel. I am now checking this with a VAR model. First I determine the optimum number of lags. According to the Schwarz criterion, two lags must be selected accord ingly. The Akaike Criterion and the Hannan-Quinn Criterion recommend more lags, but I try to avoid overfitting.12 I also look at the autocorrelation 11 The exact results of the regressions can be found in the appendix B.11 to B.14 can be found. 12 Over-adjustment is particularly problematic here, as the equity and portfolio data are only approximated to monthly data using quarterly data. In addition, this data only approximates 29 functions (ACF) and the partial autocorrelation functions (PACF) to see how the lags of the time series behave. Furthermore, statements can be made about the stationarity of the data. Stationarity is also checked using the augmented Dicky-Fuller test, as this is an important prerequisite for the application of a VAR model. In the following ACF plots we can see the autocorrelation of the different variables.
Figure 17: ACFs of wealth channel. The slowly decreasing values
indicate stationarity, this indicates that this time series can be well
described by an AR process. When checking the time series of the
variables for stationarity using the Dickey-Fuller test, all variables
with the corresponding p-value are > 0.05 and therefore the
assumption of station arity can be rejected. Since we assume
non-stationarity but want to estimate the VAR model, I now test for
cointegration to obtain a vector error corrected model (VECM), which is
applicable to non-stationary time series but requires cointegration. I
also use the Johansen test to check cointegration relationships. There
are at least two cointegration relationships at a critical value of 5%.
A VECM model is therefore used, since r > 0, namely r = 2.13 The VECM
can now be estimated. The short-term and long-term dynamics of the
variables the transmission channel, as according to Anton (2015) this
channel functions via the price of stocks and housing. 13 The results of
the Johansen test can be found in the appendix at B.15. 30 can be
determined. When analyzing the error correction terms, it is noticeable
that MRO and OMO do not have strong reactions and that consumption, in
this case final household consumption, corrects by 35% per period. When
an alyzing the short-term dynamics, the strong positive influence of OMO
and MRO on household equity and investment is striking. The influence of
equity and investments on the consumption quantities of households is
compara tively moderate.
In order to examine the shock, or the change in MPs at the end of 2022,
in more detail, I am now carrying out impulsive-response-functions based
on the VECM. The four graphs show the impulse response functions of the
wealth transmission channels. The selected time horizon is 12 periods,
i.e. one year.
It is interesting to note that in the short term, the final consumption
quantity reacts negatively to an equity and investment shock. Although
this negative shock normalizes relatively quickly after periods eight to
10, this result does not correspond to the assumed positive correlation.
The other three reaction functions correspond to the assumed positive
correlations.
31 With the VECM I was able to determine that there is a shock, but I
could not determine the structure of the shock more precisely, so I use
a structural au toregressive model (SVAR) for this. To better analyze
the shock, I use the impulsive response function of the SVAR model.
However, since no SVAR model can be estimated on the basis of a VECM
model, I estimate an SVECM model instead. The long-term effects of the
variables on each other can be seen below.14 Consumption Consumption
Equity OMO MRO 24.39 21.28 25.76 Equity 45,303.19 91.97 232.1 755.2 OMO
-0.2251 -0.0093 0.0162 400,048.05 MRO -678.97 0.00028 -0.00008 0.0011
Table 2: Coefficients of a SVECM model. 2.254 The matrix shows that a
shock in the MRO has a strong impact on consump tion and the wealth and
investment of private households. Consumption is positive in the long
term due to all variables, especially MRO. This does not speak to the
functionality of MP; the same applies to the positive influence of MSP
on the equity and investment of private households. The fact that OMO is
negatively influenced by MRO in the long term, on the other hand, is in
line with the expectations of monetary policy. The analysis using the
SVECM and the IRFs therefore relates to the entire period. If we refer
back to the figure … we can assume a difference in the effectiveness of
the transmission channel at different points in time. The fig ure … only
refers to the effects of an MRO change. I therefore assume differ ent MP
effects at different points in the time period. I proceed in a similar
way to the last transmission channel. First of all, I’m going to
identify the break points and then I’m going to identify the IRFs. This
is done in two steps, first the IRFs from equityandinvestment and
portfolio to OMO and MRO and then the response of the consumption
variables to equityandinvestment and port folio. 14 In addition to the
estimated long-term matrix, the estimated short-term matrix and the co
variance matrix of the residuals in reduced form, the other results of
the SVECM model can be found in the appendix at B.16. 32
As already assumed from the previous analysis, the influence of the variable MRO is temporally in the second half of the data set. It appears that the impact of MRO on portfolio is slightly stronger than for equityandinvestment, but relatively similar in structure. The effect of OMO is positive during the first half of the time period and negative thereafter. Of the two figures … and … Figure …. is particularly relevant here due to the higher explanatory value of equityandinvestment for the consumption variables. A shock from eq uityandinvestment has a negative impact on durable goods. The impact on final consumption reverses after half of the time period, similar to the impact of OMO on the transition variables.
4.3 Liquidity channel
For the liquidity channel that is efficient via the financial assets
of the house holds are also other channels like the signaling channel or
the unanticipated price level channel also function via an asset price –
monetary policy dy namic. The signaling channel suggested by Eggerson
and Woodford (2003) functions via the investors expectation reactions on
a monetary policy dy namic and the unanticipated price level channel is
also mentioned by Anton 33 (2015) leading via unanticipated price level
to adverse selection leading to asset price dynamics (cf. Mishkin, 2019,
p.668). In their study, Albert and Gómez-Fernández (2024) analyzed the
effects of the Fed’s monetary policy considering the assets of
households via the net worth of private American households. However,
net worth as a variable is not listed as a monthly variable for the
eurozone by either the ECB or Euro stat. Similarly, not all eurozone
countries collect data on private household net worth, so it is also not
possible to compile a data set in this way. I therefore approximate the
asset changes of households using the ECB’s monthly data on real estate
funds combined with the household net saving rate data from Eurostat.
Another possibility for measuring the monetary policies impact on the
household’s consumption is with the measurement of financial distress.
It is important to separate the effect of financial assets on financial
distress from other effects on financial distress. Other factors for
financial distress accord ing to Coughlin et al. (2021) are unemployment
and the number of hours worked. Anton (2015) measured efficiency of the
household’s liquidity chan nel with the indebtedness of the households
and the risk premium of lending for households. This results in two
regression equations for the direct channel relationship, as financial
distress cannot be measured directly via a variable. To measure
financial distress I use the consumer confidence index (CCI). The CCI
measures how households feel, optimistic or pessimistic, about their fi
nancial situation. Furthermore, there is a third regression equation to
test the efficiency of the channel based on Anton (2015): ASSETSt = α +
β1 × MROt + β2 × OMOt + β3 × realGDPt + ɛt, (8) CCIt = α + β1 × ASSETSt
+ β2 × UNEMPLOYt + β3 × GINIt + β4 × (ASSETSt × OMOt) + β5 × (ASSETSt ×
MROt) + βt × realGDPt + ɛt. (9) Consumptiont = α + β1 × CCIt + β2 ×
ASSETSt + β3 × GINIt + β4 × UNEM PLOYt + β5 × (CCIt × ASSETSt) + ɛt.
(10) The first regression shows the relationship between monetary policy
instru ments and financial assets. Due to the asset price channel, I
assume a negative correlation between MRO and financial assets and a
positive correlation 34 between OMO and financial assets. Since the
data15 on the financial assets of households are not directly available
monthly, I use the quarterly dataset and estimate the monthly data on
that basis. The total assets of the private sector and real estate data
are available monthly but are even more unprecise, be cause of the
inclusion of other actors of the economy. As companies and in
stitutional investors are also invested in real estate funds alongside
house holds, the financial assets of households are at best estimated
from quarterly data, even though of the increase in data inertia and the
reduced adaptability of the data. As a kind of control mechanism, I
nevertheless try to examine monthly data using the annual loan change of
euro area households. The as sumption here is that a household has a
certain budget, if the liquidity de creases, then the household will
sell assets and take out a loan. This would also have a negative impact
on financial distress, CCI. The consumption var iable in regression (10)
is again divided into final consumption and durable consumption of the
household. The focus is here on the final consumption of households
because of the household’s liquidity transmission channel influ ences
according to Anton (2015) the consumption of households overall.
In the following he results for the regression (8). The regression model
is ra ther simple, therefore the results are as expected. Intercept MRO
9.605e+05 Intercept OMO 2.289e+03 Sig. level MRO 0.001 Sig. level OMO
0.001 R2 0.6611 Table 3: Results of the OLS-regression of the
MP-reaction of the liquidity channel. The following table shows the
results for regression (9), where CCI is an en dogenous variable. The R2
of regression model with all control variables is 0.78 and the intercept
and significance of the variables of interest is presented in the
following table. 15 The data describe the total assets of real estate
funds in the eurozone, the marginal changes are from year to year, the
data are available in the ECB database 35 assets Intercept p-value
-3.123e-06 0.31337
Assets × MRO 3.171e-07 0.72622
Assets × OMO 2.655e-0 0.12053 Table 4: Results of the OLS-regression of
the asset-reaction of the liquidity channel. The assets do not have a
significant effect on the CCI directly, at most in combination with the
MRO. This makes sense as the CCI is relatively indi rectly affected by
asset changes and unemployment has a more direct impact on financial
distress. For the sake of completeness, I summarize below the results
for the same regressions but performed with the annual change in
household loans.
MRO OMO hsloans hsloans × MRO hsloans × OMO Intercept -3.917e 01 -4.435e
04 -2.607e+03 -1.962e+00 -3.860e-03 p-value 4.35e-09 0.114 1.54e-11
0.01180 0.03305 Table 5: Results of the OLS-regression with
interaction-terms of the 1st part of the liquidity channel. The R2 for
the first regression OMO and MRO is 0.9 and for the second re gression
with the interaction terms and the effect on CCI is 0.8. The coeffi
cients for the respective first regressions relating to the effects of
MRO on assets and hsloans are not as expected. The assets coefficients
of the MRO is positive, but expected transmission correlation is
expected negative. The OMO coefficient is as expected. Regarding the
variable hsloans the coeffi cient of the MRO variable is negative this
could be due to an increase in the interest rates for household loans, I
could already show this connection in the real interest rate channel.
The real interest rates for the household loans are rather similar to
the MRO-rates. The coefficient of the OMO variable is as expected. The
interaction terms of hsloans are all significant to the 0.01 level. The
coefficient of this interaction terms is negative, which describes a
nega tive effect in terms of the effectiveness of the MP if the loans of
the house holds increase. The interaction-terms were all positive, the
most significant is the OMO interaction term.
36 The results of the regression model (10) indicate that the
interaction between CCI and assets households is significant for both
endogenous consumption variables. Thus, the effect c.b. of CCI on
consumption increases when house hold assets increase and when
households are more pessimistic than c.b. ef fect of household assets on
consumption increases. In the following table the results of the
regression model (10) with final consumption as the endogenous variable.
The regression model has a R2 of 0.94 and an adjusted R2 of 0.93.
variable coefficient p-value assets -1.461e-02 0.34541 CCI 1.643e+05
Unemployment 1.68e-06 *** -1.061e+05 GDP 1.95e-08 *** -4.005e+05 0.00291
** Gini -2.314e+05 0.02627 * Assets × CCI -5.636e-03 1.13e-06 *** Table
6: Results of the OLS-regression with interaction-terms of the 2nd part
of the liquidity channel. If this model is estimated without the
interaction term, the explanatory power is reduced (R2)16 and the
variable becomes highly significant and the variable cci becomes
insignificant. Thus, the impact of assets on consumption would be
overestimated. This indicates that the variable assets impacts the
consump tion through the variable CCI. For durable goods the results are
similar, the signs of coefficients are the same.
As with the two previous transmission channels, I now continue the
analysis with autoregressive models to account for the temporal lags of
the effects. I run the augmented dickey fuller test with the different
variants of the trans mission channel, once with final consumption, then
once with durable goods as consumption variables and one variant where I
replace the variable assets with the variable hsloans. I also include
realGDP and unemployment as rele vant control variables. None of the
tested variables is stationary in one data set, the p-value > 0.05
and therefore each of the variables in the three data sets must be
rejected from the assumption of stationarity.17
16 The R2 for the model without interaction term is 0.9. 17 The detailed
results of the augmented dickey fuller tests of the variables can be
found in the appendix
37 Now I am testing the variables for cointegration using the Johansen
test. There are at least five long-term cointegration relationships
between the variables. Therefore, a VECM is estimated first and not a
VAR. The VECM model re veals some significant correlations. The assets,
which previously had no sig nificant influence on CCI, now had a
significant positive influence two lags earlier. Furthermore, the
multiple determinants of assets become clear. Not only do past MRO
values have a negative and past OMO values a positive significant
influence on the assets, but past values of final consumption and CCI
also have a significant positive influence. Contrary to the results of
the OLS regression (8), the results of the VECM with final consumption
as a consumption variable show a negative coefficient for MRO. The
coefficient of MRO becomes more negative the further back in time,
i.e. the higher the lag, the more negative the coefficient of MRO (up to
the fifth lag). In addition to the variable itself in the past, CCI with
lag(2) also has a significant positive influence on durable goods. If
the variable hsloans is used for the variable assets, the dynamics do
not change fundamentally. OMO has no influence on the variable hsloans,
but MRO has a significant negative influence at different lags.
Now I analyze the reactions of the variables in more detail and use IRFs
for this. The time series of the assets variable has been differentiated
beforehand for better presentation in the IRF.
##
## Call:
## lm(formula = dataeurozone$hsassets ~ dataeurozone$MRO + dataeurozone$OMO +
## dataeurozone$GDP, data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1505512 -601564 -51995 636986 1553711
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.999e+07 7.494e+06 5.336 2.42e-06 ***
## dataeurozone$MRO 9.605e+05 1.023e+05 9.392 1.55e-12 ***
## dataeurozone$OMO 2.289e+03 5.127e+02 4.466 4.70e-05 ***
## dataeurozone$GDP -1.046e+07 5.795e+06 -1.806 0.0771 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 826300 on 49 degrees of freedom
## (7 observations deleted due to missingness)
## Multiple R-squared: 0.6611, Adjusted R-squared: 0.6404
## F-statistic: 31.87 on 3 and 49 DF, p-value: 1.425e-11##
## Call:
## lm(formula = dataeurozone$CCI ~ dataeurozone$hsassets + dataeurozone$unemployment +
## dataeurozone$GDP + dataeurozone$Gini + (dataeurozone$hsassets *
## dataeurozone$MRO) + (dataeurozone$hsassets * dataeurozone$OMO),
## data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.8399 -2.3182 0.1287 2.5750 5.8893
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.993e+03 6.830e+02 -2.917 0.00554
## dataeurozone$hsassets -3.123e-06 3.063e-06 -1.020 0.31337
## dataeurozone$unemployment 2.759e-01 1.966e+00 0.140 0.88902
## dataeurozone$GDP -8.107e+01 3.103e+01 -2.613 0.01225
## dataeurozone$Gini 3.011e+01 9.405e+00 3.202 0.00254
## dataeurozone$MRO -9.514e+00 2.606e+01 -0.365 0.71685
## dataeurozone$OMO -7.412e-02 4.788e-02 -1.548 0.12877
## dataeurozone$hsassets:dataeurozone$MRO 3.171e-07 8.999e-07 0.352 0.72622
## dataeurozone$hsassets:dataeurozone$OMO 2.655e-09 1.677e-09 1.583 0.12053
##
## (Intercept) **
## dataeurozone$hsassets
## dataeurozone$unemployment
## dataeurozone$GDP *
## dataeurozone$Gini **
## dataeurozone$MRO
## dataeurozone$OMO
## dataeurozone$hsassets:dataeurozone$MRO
## dataeurozone$hsassets:dataeurozone$OMO
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.361 on 44 degrees of freedom
## (7 observations deleted due to missingness)
## Multiple R-squared: 0.7818, Adjusted R-squared: 0.7421
## F-statistic: 19.7 on 8 and 44 DF, p-value: 3.262e-12##
## Call:
## lm(formula = dataeurozone$finalconsumption ~ dataeurozone$hsassets +
## dataeurozone$CCI + dataeurozone$unemployment + dataeurozone$GDP +
## dataeurozone$Gini + (dataeurozone$hsassets * dataeurozone$CCI),
## data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -173612 -22129 10297 29353 68448
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.015e+07 7.153e+06 2.817 0.00693
## dataeurozone$hsassets -1.461e-02 1.533e-02 -0.953 0.34541
## dataeurozone$CCI 1.643e+05 3.028e+04 5.426 1.68e-06
## dataeurozone$unemployment -1.061e+05 1.590e+04 -6.672 1.95e-08
## dataeurozone$GDP -4.005e+05 1.279e+05 -3.131 0.00291
## dataeurozone$Gini -2.314e+05 1.011e+05 -2.290 0.02627
## dataeurozone$hsassets:dataeurozone$CCI -5.636e-03 1.018e-03 -5.539 1.13e-06
##
## (Intercept) **
## dataeurozone$hsassets
## dataeurozone$CCI ***
## dataeurozone$unemployment ***
## dataeurozone$GDP **
## dataeurozone$Gini *
## dataeurozone$hsassets:dataeurozone$CCI ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 47320 on 50 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.9437, Adjusted R-squared: 0.9369
## F-statistic: 139.6 on 6 and 50 DF, p-value: < 2.2e-16##
## Call:
## lm(formula = dataeurozone$durablehs ~ dataeurozone$hsassets +
## dataeurozone$CCI + dataeurozone$unemployment + dataeurozone$GDP +
## dataeurozone$Gini + (dataeurozone$hsassets * dataeurozone$CCI),
## data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.80958 -0.64433 -0.06754 0.68149 1.76781
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 8.560e+01 2.881e+02 0.297 0.76891
## dataeurozone$hsassets -2.239e-06 9.551e-07 -2.345 0.02766
## dataeurozone$CCI 8.170e+00 2.255e+00 3.623 0.00136
## dataeurozone$unemployment -1.757e+00 1.307e+00 -1.344 0.19144
## dataeurozone$GDP 3.625e+00 6.655e+00 0.545 0.59106
## dataeurozone$Gini -1.595e-01 3.906e+00 -0.041 0.96777
## dataeurozone$hsassets:dataeurozone$CCI -2.711e-07 7.628e-08 -3.555 0.00161
##
## (Intercept)
## dataeurozone$hsassets *
## dataeurozone$CCI **
## dataeurozone$unemployment
## dataeurozone$GDP
## dataeurozone$Gini
## dataeurozone$hsassets:dataeurozone$CCI **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.039 on 24 degrees of freedom
## (29 observations deleted due to missingness)
## Multiple R-squared: 0.8878, Adjusted R-squared: 0.8597
## F-statistic: 31.64 on 6 and 24 DF, p-value: 2.923e-10##
## Call:
## lm(formula = dataeurozone$loanshsannualgrowth ~ dataeurozone$MRO +
## dataeurozone$OMO + dataeurozone$GDP, data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.72141 -0.22386 0.00052 0.13863 1.29383
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -2.411e+01 4.032e+00 -5.980 2.52e-07 ***
## dataeurozone$MRO -3.917e-01 5.503e-02 -7.118 4.35e-09 ***
## dataeurozone$OMO -4.435e-04 2.758e-04 -1.608 0.114
## dataeurozone$GDP 2.013e+01 3.118e+00 6.456 4.62e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4446 on 49 degrees of freedom
## (7 observations deleted due to missingness)
## Multiple R-squared: 0.9077, Adjusted R-squared: 0.902
## F-statistic: 160.6 on 3 and 49 DF, p-value: < 2.2e-16##
## Call:
## lm(formula = dataeurozone$CCI ~ dataeurozone$loanshsannualgrowth +
## dataeurozone$unemployment + dataeurozone$GDP + dataeurozone$Gini +
## (dataeurozone$loanshsannualgrowth * dataeurozone$MRO) + (dataeurozone$loanshsannualgrowth *
## dataeurozone$OMO), data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.4457 -1.9699 -0.3717 2.2380 6.0074
##
## Coefficients:
## Estimate Std. Error t value
## (Intercept) -2.607e+03 2.896e+02 -9.002
## dataeurozone$loanshsannualgrowth 1.182e+01 3.672e+00 3.217
## dataeurozone$unemployment 2.898e-01 1.722e+00 0.168
## dataeurozone$GDP -8.368e+01 3.253e+01 -2.572
## dataeurozone$Gini 3.687e+01 4.087e+00 9.021
## dataeurozone$MRO 6.573e+00 2.521e+00 2.607
## dataeurozone$OMO 1.574e-02 6.882e-03 2.287
## dataeurozone$loanshsannualgrowth:dataeurozone$MRO -1.962e+00 7.466e-01 -2.627
## dataeurozone$loanshsannualgrowth:dataeurozone$OMO -3.860e-03 1.754e-03 -2.201
## Pr(>|t|)
## (Intercept) 1.54e-11 ***
## dataeurozone$loanshsannualgrowth 0.00243 **
## dataeurozone$unemployment 0.86713
## dataeurozone$GDP 0.01357 *
## dataeurozone$Gini 1.45e-11 ***
## dataeurozone$MRO 0.01242 *
## dataeurozone$OMO 0.02709 *
## dataeurozone$loanshsannualgrowth:dataeurozone$MRO 0.01180 *
## dataeurozone$loanshsannualgrowth:dataeurozone$OMO 0.03305 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.193 on 44 degrees of freedom
## (7 observations deleted due to missingness)
## Multiple R-squared: 0.8029, Adjusted R-squared: 0.7671
## F-statistic: 22.41 on 8 and 44 DF, p-value: 3.727e-13##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf1$MRO
## Dickey-Fuller = -2.7817, Lag order = 3, p-value = 0.2598
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf1$OMO
## Dickey-Fuller = -2.4946, Lag order = 3, p-value = 0.3754
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf1$hsassets
## Dickey-Fuller = -1.8033, Lag order = 3, p-value = 0.6536
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf1$CCI
## Dickey-Fuller = -2.2069, Lag order = 3, p-value = 0.4912
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf1$finalconsumption
## Dickey-Fuller = -2.8751, Lag order = 3, p-value = 0.2222
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf2$MRO
## Dickey-Fuller = -0.050588, Lag order = 2, p-value = 0.99
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf2$OMO
## Dickey-Fuller = -1.5772, Lag order = 2, p-value = 0.7339
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf2$hsassets
## Dickey-Fuller = -3.3117, Lag order = 2, p-value = 0.08966
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf2$CCI
## Dickey-Fuller = -3.0328, Lag order = 2, p-value = 0.1782
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf2$durablehs
## Dickey-Fuller = -2.7, Lag order = 2, p-value = 0.3052
## alternative hypothesis: stationaryVariable assets react positively to a shock in the MRO as well as the
OMO after two to four months. The response to the variable OMO is
positive from B.17. 38 period one, while the response of the variable is
initially negative in the first period and then positive from the second
period onwards. The fact that vari able assets initially reacted
negatively to the MRO shock in the first period is consistent with the
results of the VECM. In the IRF of the figure we see that the short-term
effect of a shock is negative, but there could be long-term ad justment
mechanisms via third variables. These adjustment mechanisms then lead to
a positive effect. If we look at the long-term beta coefficients of the
VECM, the coefficient of MRO and assets is slightly positive. For OMO,
on the other hand, it is slightly negative, which is also shown in
Figure 27.18
The transmission works here via the assets, which then influence consump
tion via the financial distress. I assume a positive correlation between
the var iables; this has already been shown in the context of the VECM
coefficients. In the figure 28 shows a positive reaction to a shock in
the asset variable. The reaction of final consumption to CCI is similar.
The effect of a positive shock to the propensity to consume is more
short-term than the positive reaction to a shock to household
assets.
##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf3$MRO
## Dickey-Fuller = -2.7817, Lag order = 3, p-value = 0.2598
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf3$OMO
## Dickey-Fuller = -2.4946, Lag order = 3, p-value = 0.3754
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf3$loanshsannualgrowth
## Dickey-Fuller = -2.6422, Lag order = 3, p-value = 0.316
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf3$CCI
## Dickey-Fuller = -2.2069, Lag order = 3, p-value = 0.4912
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: data_subsetadf3$finalconsumption
## Dickey-Fuller = -2.8751, Lag order = 3, p-value = 0.2222
## alternative hypothesis: stationaryThe two consumption variables, final consumption and durable goods,
differ in their effect on the response to the. Durable goods react more
volatile to a shock in CCI as the time period increases. This may
indicate different things, from an overreaction of durable goods to CCI
or persistence of the shock, but for statistical problem of the time
series. In figure 29 also shows the IRF of CCI on the change in
household loans. 18 The long-term beta coefficient of MRO on assets is
3.242597e-06 and of OMO on assets 1.283621e-06. 39 Figure 30: IRF of CCI
to hsloans. Figure 31: IRF of durable goods con sumption to CCI. The
shock reaction of CCI is comparable to that of the assets variable. The
IRF of the Figure 30 looks slightly different at first, but hsloan
variable annual changes are measured and the assets variable absolute
values, if this is taken into account, then the reaction of CCI is
relatively similar.
In order to look at the reactions to shocks of the variable within the
transmis sion channel as a whole, the breakpoints of the time series are
now included, similar to the previous two channels. The MRO and OMO
breakpoints are simultaneous to the breakpoints of the MRO and OMO last
transmission channel. In the best model, OMO has four breakpoints in
time periods 7, 14, 34, and 41 and MRO at time periods 32 and 40.
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 6 6 6 7
##
## $criteria
## 1 2 3 4 5 6
## AIC(n) 3.667078e+01 3.516962e+01 3.453126e+01 3.189679e+01 2.078840e+01 -Inf
## HQ(n) 3.751661e+01 3.675555e+01 3.685729e+01 3.496292e+01 2.459463e+01 -Inf
## SC(n) 3.896443e+01 3.947022e+01 4.083882e+01 4.021129e+01 3.110985e+01 -Inf
## FPE(n) 8.694516e+15 2.327488e+15 2.093314e+15 5.376284e+14 2.021413e+11 NaN
## 7 8 9 10
## AIC(n) -Inf -Inf -Inf -Inf
## HQ(n) -Inf -Inf -Inf -Inf
## SC(n) -Inf -Inf -Inf -Inf
## FPE(n) 0 0 0 0## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 7 7 7 8
##
## $criteria
## 1 2 3 4 5
## AIC(n) 4.577159e+01 4.426627e+01 4.409700e+01 4.268494e+01 4.153193e+01
## HQ(n) 4.640596e+01 4.544439e+01 4.581886e+01 4.495056e+01 4.434129e+01
## SC(n) 4.749183e+01 4.746101e+01 4.876622e+01 4.882866e+01 4.915015e+01
## FPE(n) 7.690667e+19 1.885470e+19 2.096011e+19 9.360338e+18 1.047705e+19
## 6 7 8 9 10
## AIC(n) 3.684674e+01 -Inf -Inf -Inf -Inf
## HQ(n) 4.019985e+01 -Inf -Inf -Inf -Inf
## SC(n) 4.593944e+01 -Inf -Inf -Inf -Inf
## FPE(n) 1.851926e+18 NaN 0 0 0##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 1.00000003 1.00000000 1.00000000 0.81769898 0.68063415 0.02324985
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 5 | 1.08 6.50 8.18 11.65
## r <= 4 | 53.59 15.66 17.95 23.52
## r <= 3 | 131.88 28.71 31.52 37.22
## r <= 2 | 1015.12 45.23 48.28 55.43
## r <= 1 | 2031.13 66.49 70.60 78.87
## r = 0 | NaN 85.18 90.39 104.20
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## MRO.l7 OMO.l7 unemployment.l7 hsassets.l7
## MRO.l7 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## OMO.l7 -1.357998e-03 2.084970e-03 1.010085e-04 1.466144e-03
## unemployment.l7 -3.848118e+00 2.971941e+00 -2.751053e+01 -1.491199e+01
## hsassets.l7 3.242597e-06 -1.283622e-06 3.094286e-07 4.944256e-06
## CCI.l7 -3.234463e-01 9.149096e-02 -2.853877e-01 -7.118838e-01
## finalconsumption.l7 -5.034628e-05 1.219657e-05 -1.391530e-04 -1.007113e-04
## CCI.l7 finalconsumption.l7
## MRO.l7 1.000000e+00 1.000000e+00
## OMO.l7 4.772208e-04 1.735448e-03
## unemployment.l7 -2.576636e+00 -1.812226e+00
## hsassets.l7 1.240569e-06 2.146936e-07
## CCI.l7 -1.853375e-01 -7.829034e-02
## finalconsumption.l7 -2.756593e-05 -1.328322e-05
##
## Weights W:
## (This is the loading matrix)
##
## MRO.l7 OMO.l7 unemployment.l7 hsassets.l7
## MRO.d -3.181668e-01 -2.977190e-01 -6.283346e-03 -3.293779e-02
## OMO.d 2.710819e+01 9.623764e+01 2.287982e+01 1.041839e+02
## unemployment.d 3.064224e-01 -3.991183e-02 5.296865e-02 -1.193265e-01
## hsassets.d -5.569704e+05 1.331239e+05 4.497503e+03 -7.979305e+04
## CCI.d -1.286295e+01 3.961003e+00 -5.498885e-01 2.321421e-01
## finalconsumption.d 1.370043e+04 5.481226e+03 -8.757661e+02 4.792186e+02
## CCI.l7 finalconsumption.l7
## MRO.d 9.229821e-01 -1.478462e-01
## OMO.d -5.457273e+01 9.157596e+01
## unemployment.d 6.656479e-01 5.464623e-02
## hsassets.d -5.850289e+05 1.658796e+04
## CCI.d 2.806338e+01 -1.350990e+00
## finalconsumption.d -5.253171e+04 -8.806050e+02##
## Call:
## lm(formula = substitute(form1), data = data.mat)
##
## Coefficients:
## MRO.d OMO.d unemployment.d hsassets.d
## ect1 2.679e-01 1.958e+02 8.658e-01 -1.084e+06
## ect2 2.029e-04 2.929e-01 -3.513e-04 6.382e+02
## ect3 -1.375e+00 -1.861e+03 -2.691e+00 5.112e+06
## ect4 3.307e-07 4.189e-04 1.297e-06 -3.096e+00
## ect5 -7.015e-02 -7.055e+01 -1.563e-01 3.563e+05
## constant 1.364e+01 2.111e+04 2.846e+01 -2.945e+07
## MRO.dl1 -5.399e-01 -9.222e+02 3.137e-01 -1.703e+03
## OMO.dl1 5.719e-04 -1.261e+00 -4.956e-04 2.122e+02
## unemployment.dl1 -3.570e-01 -1.797e+02 -1.163e+00 4.776e+05
## hsassets.dl1 -1.608e-07 2.310e-04 4.026e-07 -6.516e-01
## CCI.dl1 3.650e-02 2.649e-01 -2.511e-02 2.862e+04
## finalconsumption.dl1 -1.089e-05 -1.368e-02 -1.333e-05 2.649e+01
## MRO.dl2 7.058e-02 -1.318e+03 3.131e-01 -6.196e+05
## OMO.dl2 1.043e-03 -8.363e-01 -8.151e-04 -5.561e+02
## unemployment.dl2 -8.528e-01 -4.543e+01 -1.259e+00 1.407e+06
## hsassets.dl2 6.310e-07 -2.291e-04 -1.227e-07 -1.023e+00
## CCI.dl2 1.409e-02 3.044e+01 -5.089e-02 6.987e+04
## finalconsumption.dl2 1.859e-06 -1.322e-02 -1.591e-05 1.618e+01
## MRO.dl3 4.676e-01 -9.297e+02 6.193e-01 -1.499e+06
## OMO.dl3 3.947e-05 1.580e-01 -4.270e-05 2.325e+02
## unemployment.dl3 -4.573e-01 -5.477e+02 -1.330e+00 2.487e+06
## hsassets.dl3 -7.999e-07 6.543e-04 1.104e-06 -1.885e+00
## CCI.dl3 1.985e-02 5.914e+01 -6.348e-02 1.192e+05
## finalconsumption.dl3 -5.014e-06 -2.835e-03 -7.948e-06 1.698e+01
## MRO.dl4 2.539e-01 -3.096e+02 1.188e+00 -1.911e+06
## OMO.dl4 5.834e-04 -5.679e-01 -3.297e-04 6.887e+02
## unemployment.dl4 -1.221e+00 -6.058e+02 -1.884e+00 3.938e+06
## hsassets.dl4 7.519e-07 -3.335e-04 1.124e-06 -1.875e+00
## CCI.dl4 -3.959e-02 6.259e+01 -8.606e-02 1.980e+05
## finalconsumption.dl4 -5.124e-06 -1.472e-02 -2.044e-05 3.483e+01
## MRO.dl5 4.424e-01 -2.565e+01 1.522e+00 -2.067e+06
## OMO.dl5 4.735e-04 -5.068e-01 -3.104e-05 1.098e+03
## unemployment.dl5 -7.962e-01 -1.601e+03 -2.547e+00 4.501e+06
## hsassets.dl5 3.239e-07 -5.107e-04 2.720e-07 -2.239e+00
## CCI.dl5 -4.319e-02 1.338e+01 -1.105e-01 2.759e+05
## finalconsumption.dl5 -4.324e-06 -9.442e-03 -1.836e-05 3.578e+01
## MRO.dl6 4.582e-01 2.664e+02 1.240e+00 -1.540e+06
## OMO.dl6 9.934e-04 -4.060e-01 -8.234e-04 1.136e+03
## unemployment.dl6 -1.851e+00 -1.402e+03 -2.897e+00 5.112e+06
## hsassets.dl6 1.688e-07 2.156e-04 2.599e-07 -3.231e+00
## CCI.dl6 -6.854e-02 -3.148e+01 -1.575e-01 3.324e+05
## finalconsumption.dl6 -1.635e-06 -7.889e-03 -2.125e-05 4.220e+01
## CCI.d finalconsumption.d
## ect1 1.884e+01 -3.375e+04
## ect2 3.940e-02 -3.163e+01
## ect3 6.269e-01 1.159e+05
## ect4 -1.100e-05 -2.568e-02
## ect5 -6.867e-01 5.715e+03
## constant 2.747e+02 -1.443e+06
## MRO.dl1 2.950e+00 2.451e+04
## OMO.dl1 -5.121e-03 3.849e+01
## unemployment.dl1 -1.140e+00 1.430e+04
## hsassets.dl1 -1.366e-06 -1.754e-02
## CCI.dl1 -1.071e+00 2.062e+02
## finalconsumption.dl1 1.062e-04 8.166e-01
## MRO.dl2 5.268e+00 4.791e+04
## OMO.dl2 -3.012e-03 2.526e+01
## unemployment.dl2 -1.510e+01 2.162e+04
## hsassets.dl2 2.338e-05 1.165e-02
## CCI.dl2 -1.352e+00 3.310e+02
## finalconsumption.dl2 -5.632e-07 8.086e-01
## MRO.dl3 -4.484e+00 2.520e+04
## OMO.dl3 -6.643e-03 -2.606e+01
## unemployment.dl3 -1.073e+01 3.808e+04
## hsassets.dl3 -9.949e-06 -2.651e-02
## CCI.dl3 -1.614e+00 -3.932e+02
## finalconsumption.dl3 -1.571e-04 -3.959e-01
## MRO.dl4 -9.583e+00 -3.414e+03
## OMO.dl4 1.343e-02 -2.046e+00
## unemployment.dl4 -6.864e+00 5.065e+04
## hsassets.dl4 1.404e-06 -2.618e-02
## CCI.dl4 -9.773e-01 2.983e+01
## finalconsumption.dl4 -8.648e-05 8.864e-01
## MRO.dl5 -1.602e+01 -1.971e+04
## OMO.dl5 1.396e-02 -9.934e+00
## unemployment.dl5 1.961e+00 7.937e+04
## hsassets.dl5 6.614e-06 4.471e-03
## CCI.dl5 -8.820e-01 1.501e+03
## finalconsumption.dl5 -1.921e-04 6.396e-01
## MRO.dl6 8.112e-01 -3.082e+04
## OMO.dl6 4.174e-02 1.985e+00
## unemployment.dl6 -1.503e+00 1.111e+05
## hsassets.dl6 -1.305e-05 -1.414e-02
## CCI.dl6 -6.192e-02 3.915e+03
## finalconsumption.dl6 -1.015e-04 2.298e-01## ect1 ect2 ect3 ect4
## MRO.l7 1.000000e+00 -4.547474e-13 0.000000e+00 0.00000
## OMO.l7 2.439455e-19 1.000000e+00 -1.084202e-19 0.00000
## unemployment.l7 2.220446e-16 0.000000e+00 1.000000e+00 0.00000
## hsassets.l7 -7.940934e-22 -1.301043e-18 3.176374e-22 1.00000
## CCI.l7 4.163336e-17 5.684342e-14 -1.387779e-17 0.00000
## finalconsumption.l7 -1.268916e-05 -1.703057e-03 5.649001e-06 -17.71491
## ect5
## MRO.l7 0.000000e+00
## OMO.l7 1.040834e-17
## unemployment.l7 0.000000e+00
## hsassets.l7 -4.065758e-20
## CCI.l7 1.000000e+00
## finalconsumption.l7 -1.212272e-04## MRO.d OMO.d unemployment.d hsassets.d
## ect1 2.678752e-01 1.958369e+02 8.658007e-01 -1.084171e+06
## ect2 2.028745e-04 2.928562e-01 -3.512743e-04 6.382021e+02
## ect3 -1.374623e+00 -1.860715e+03 -2.690698e+00 5.112476e+06
## ect4 3.306972e-07 4.188587e-04 1.297027e-06 -3.095806e+00
## ect5 -7.015090e-02 -7.054526e+01 -1.563023e-01 3.562773e+05
## constant 1.364330e+01 2.111185e+04 2.846027e+01 -2.944622e+07
## MRO.dl1 -5.399185e-01 -9.221898e+02 3.137118e-01 -1.703077e+03
## OMO.dl1 5.719372e-04 -1.261255e+00 -4.955692e-04 2.121748e+02
## unemployment.dl1 -3.570071e-01 -1.796598e+02 -1.162735e+00 4.775559e+05
## hsassets.dl1 -1.607750e-07 2.309536e-04 4.025630e-07 -6.516050e-01
## CCI.dl1 3.649999e-02 2.648641e-01 -2.511498e-02 2.862470e+04
## finalconsumption.dl1 -1.089056e-05 -1.368017e-02 -1.332611e-05 2.649107e+01
## MRO.dl2 7.058122e-02 -1.317886e+03 3.130824e-01 -6.195839e+05
## OMO.dl2 1.042840e-03 -8.362560e-01 -8.151242e-04 -5.560657e+02
## unemployment.dl2 -8.527730e-01 -4.543291e+01 -1.259001e+00 1.406756e+06
## hsassets.dl2 6.310350e-07 -2.291473e-04 -1.227200e-07 -1.023195e+00
## CCI.dl2 1.409149e-02 3.044138e+01 -5.088823e-02 6.987247e+04
## finalconsumption.dl2 1.858994e-06 -1.321905e-02 -1.590541e-05 1.618300e+01
## MRO.dl3 4.676247e-01 -9.296728e+02 6.193367e-01 -1.499425e+06
## OMO.dl3 3.947216e-05 1.579519e-01 -4.269770e-05 2.325015e+02
## unemployment.dl3 -4.573182e-01 -5.477454e+02 -1.330146e+00 2.486660e+06
## hsassets.dl3 -7.999046e-07 6.542786e-04 1.104072e-06 -1.885310e+00
## CCI.dl3 1.984612e-02 5.914140e+01 -6.347677e-02 1.191876e+05
## finalconsumption.dl3 -5.014056e-06 -2.835178e-03 -7.947667e-06 1.697556e+01
## MRO.dl4 2.539380e-01 -3.095589e+02 1.188131e+00 -1.910969e+06
## OMO.dl4 5.833609e-04 -5.679039e-01 -3.296732e-04 6.887198e+02
## unemployment.dl4 -1.221193e+00 -6.058441e+02 -1.884040e+00 3.938027e+06
## hsassets.dl4 7.518501e-07 -3.334774e-04 1.124208e-06 -1.874550e+00
## CCI.dl4 -3.958710e-02 6.258616e+01 -8.606236e-02 1.980112e+05
## finalconsumption.dl4 -5.124217e-06 -1.471639e-02 -2.044239e-05 3.483021e+01
## MRO.dl5 4.424238e-01 -2.565010e+01 1.521978e+00 -2.066630e+06
## OMO.dl5 4.734551e-04 -5.068495e-01 -3.103860e-05 1.097556e+03
## unemployment.dl5 -7.961516e-01 -1.601352e+03 -2.547421e+00 4.501185e+06
## hsassets.dl5 3.238818e-07 -5.107363e-04 2.719676e-07 -2.238803e+00
## CCI.dl5 -4.319196e-02 1.338262e+01 -1.104526e-01 2.759127e+05
## finalconsumption.dl5 -4.324185e-06 -9.442054e-03 -1.835656e-05 3.577827e+01
## MRO.dl6 4.581814e-01 2.664444e+02 1.240179e+00 -1.540207e+06
## OMO.dl6 9.934325e-04 -4.059904e-01 -8.233949e-04 1.136498e+03
## unemployment.dl6 -1.850928e+00 -1.401856e+03 -2.896569e+00 5.111752e+06
## hsassets.dl6 1.688109e-07 2.155935e-04 2.598578e-07 -3.230685e+00
## CCI.dl6 -6.853737e-02 -3.147981e+01 -1.575447e-01 3.323720e+05
## finalconsumption.dl6 -1.635330e-06 -7.888760e-03 -2.125471e-05 4.219882e+01
## CCI.d finalconsumption.d
## ect1 1.884369e+01 -3.374660e+04
## ect2 3.940366e-02 -3.163203e+01
## ect3 6.269022e-01 1.158708e+05
## ect4 -1.100162e-05 -2.568164e-02
## ect5 -6.866568e-01 5.715013e+03
## constant 2.746985e+02 -1.442859e+06
## MRO.dl1 2.950320e+00 2.450954e+04
## OMO.dl1 -5.120858e-03 3.848972e+01
## unemployment.dl1 -1.140176e+00 1.430311e+04
## hsassets.dl1 -1.365596e-06 -1.753856e-02
## CCI.dl1 -1.070667e+00 2.061721e+02
## finalconsumption.dl1 1.062019e-04 8.165663e-01
## MRO.dl2 5.267681e+00 4.790901e+04
## OMO.dl2 -3.012493e-03 2.526283e+01
## unemployment.dl2 -1.510090e+01 2.161955e+04
## hsassets.dl2 2.338056e-05 1.165364e-02
## CCI.dl2 -1.352002e+00 3.309761e+02
## finalconsumption.dl2 -5.631785e-07 8.085842e-01
## MRO.dl3 -4.484139e+00 2.519757e+04
## OMO.dl3 -6.642792e-03 -2.606217e+01
## unemployment.dl3 -1.073329e+01 3.808482e+04
## hsassets.dl3 -9.948986e-06 -2.651382e-02
## CCI.dl3 -1.613900e+00 -3.931783e+02
## finalconsumption.dl3 -1.570992e-04 -3.958733e-01
## MRO.dl4 -9.583466e+00 -3.414270e+03
## OMO.dl4 1.342776e-02 -2.045779e+00
## unemployment.dl4 -6.863628e+00 5.064894e+04
## hsassets.dl4 1.404095e-06 -2.617652e-02
## CCI.dl4 -9.772713e-01 2.982564e+01
## finalconsumption.dl4 -8.648416e-05 8.863941e-01
## MRO.dl5 -1.602293e+01 -1.971082e+04
## OMO.dl5 1.396003e-02 -9.934126e+00
## unemployment.dl5 1.960974e+00 7.936604e+04
## hsassets.dl5 6.614073e-06 4.470988e-03
## CCI.dl5 -8.819742e-01 1.500781e+03
## finalconsumption.dl5 -1.921095e-04 6.395694e-01
## MRO.dl6 8.111801e-01 -3.082186e+04
## OMO.dl6 4.174185e-02 1.984815e+00
## unemployment.dl6 -1.502800e+00 1.111454e+05
## hsassets.dl6 -1.305457e-05 -1.414281e-02
## CCI.dl6 -6.191824e-02 3.915394e+03
## finalconsumption.dl6 -1.014790e-04 2.297971e-01## Response MRO.d :
##
## Call:
## lm(formula = MRO.d ~ ect1 + ect2 + ect3 + ect4 + ect5 + constant +
## MRO.dl1 + OMO.dl1 + unemployment.dl1 + hsassets.dl1 + CCI.dl1 +
## finalconsumption.dl1 + MRO.dl2 + OMO.dl2 + unemployment.dl2 +
## hsassets.dl2 + CCI.dl2 + finalconsumption.dl2 + MRO.dl3 +
## OMO.dl3 + unemployment.dl3 + hsassets.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + unemployment.dl4 + hsassets.dl4 + CCI.dl4 +
## finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + unemployment.dl5 +
## hsassets.dl5 + CCI.dl5 + finalconsumption.dl5 + MRO.dl6 +
## OMO.dl6 + unemployment.dl6 + hsassets.dl6 + CCI.dl6 + finalconsumption.dl6 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7 8
## 0.005216 0.003479 -0.025451 0.028801 0.008360 -0.045557 0.030739 0.014990
## 9 10 11 12 13 14 15 16
## -0.039175 0.033242 0.008382 -0.003152 -0.045926 -0.014510 0.112510 -0.051808
## 17 18 19 20 21 22 23 24
## -0.029911 0.019247 0.014182 -0.023970 0.013850 -0.017488 -0.020981 0.004321
## 25 26 27 28 29 30 31 32
## 0.032676 0.007599 -0.059545 0.020001 0.037837 -0.001426 -0.006901 -0.017225
## 33 34 35 36 37 38 39 40
## 0.018498 -0.016739 0.039650 -0.060561 0.015680 0.034273 -0.013283 -0.002476
## 41 42 43 44 45 46
## -0.032566 0.032354 -0.003820 -0.035601 -0.011286 0.043471
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 2.679e-01 7.669e-01 0.349 0.7445
## ect2 2.029e-04 7.365e-04 0.275 0.7966
## ect3 -1.375e+00 2.561e+00 -0.537 0.6200
## ect4 3.307e-07 1.734e-06 0.191 0.8580
## ect5 -7.015e-02 1.929e-01 -0.364 0.7345
## constant 1.364e+01 1.802e+01 0.757 0.4911
## MRO.dl1 -5.399e-01 5.459e-01 -0.989 0.3786
## OMO.dl1 5.719e-04 6.411e-04 0.892 0.4227
## unemployment.dl1 -3.570e-01 4.035e-01 -0.885 0.4263
## hsassets.dl1 -1.608e-07 7.537e-07 -0.213 0.8415
## CCI.dl1 3.650e-02 1.879e-02 1.943 0.1239
## finalconsumption.dl1 -1.089e-05 1.172e-05 -0.929 0.4053
## MRO.dl2 7.058e-02 9.564e-01 0.074 0.9447
## OMO.dl2 1.043e-03 4.625e-04 2.255 0.0872 .
## unemployment.dl2 -8.528e-01 6.606e-01 -1.291 0.2663
## hsassets.dl2 6.310e-07 6.067e-07 1.040 0.3570
## CCI.dl2 1.409e-02 5.197e-02 0.271 0.7997
## finalconsumption.dl2 1.859e-06 1.135e-05 0.164 0.8778
## MRO.dl3 4.676e-01 1.299e+00 0.360 0.7370
## OMO.dl3 3.947e-05 4.853e-04 0.081 0.9391
## unemployment.dl3 -4.573e-01 1.204e+00 -0.380 0.7233
## hsassets.dl3 -7.999e-07 9.729e-07 -0.822 0.4571
## CCI.dl3 1.985e-02 8.515e-02 0.233 0.8271
## finalconsumption.dl3 -5.014e-06 6.911e-06 -0.726 0.5083
## MRO.dl4 2.539e-01 1.300e+00 0.195 0.8546
## OMO.dl4 5.834e-04 7.452e-04 0.783 0.4775
## unemployment.dl4 -1.221e+00 1.844e+00 -0.662 0.5441
## hsassets.dl4 7.519e-07 1.214e-06 0.619 0.5693
## CCI.dl4 -3.959e-02 1.347e-01 -0.294 0.7835
## finalconsumption.dl4 -5.124e-06 1.870e-05 -0.274 0.7976
## MRO.dl5 4.424e-01 1.266e+00 0.349 0.7444
## OMO.dl5 4.735e-04 6.895e-04 0.687 0.5300
## unemployment.dl5 -7.962e-01 2.358e+00 -0.338 0.7527
## hsassets.dl5 3.239e-07 1.195e-06 0.271 0.7997
## CCI.dl5 -4.319e-02 1.566e-01 -0.276 0.7963
## finalconsumption.dl5 -4.324e-06 1.921e-05 -0.225 0.8329
## MRO.dl6 4.582e-01 1.033e+00 0.443 0.6804
## OMO.dl6 9.934e-04 9.409e-04 1.056 0.3506
## unemployment.dl6 -1.851e+00 2.617e+00 -0.707 0.5184
## hsassets.dl6 1.688e-07 1.697e-06 0.099 0.9256
## CCI.dl6 -6.854e-02 1.832e-01 -0.374 0.7273
## finalconsumption.dl6 -1.635e-06 2.296e-05 -0.071 0.9466
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1095 on 4 degrees of freedom
## Multiple R-squared: 0.9798, Adjusted R-squared: 0.7679
## F-statistic: 4.624 on 42 and 4 DF, p-value: 0.07208
##
##
## Response OMO.d :
##
## Call:
## lm(formula = OMO.d ~ ect1 + ect2 + ect3 + ect4 + ect5 + constant +
## MRO.dl1 + OMO.dl1 + unemployment.dl1 + hsassets.dl1 + CCI.dl1 +
## finalconsumption.dl1 + MRO.dl2 + OMO.dl2 + unemployment.dl2 +
## hsassets.dl2 + CCI.dl2 + finalconsumption.dl2 + MRO.dl3 +
## OMO.dl3 + unemployment.dl3 + hsassets.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + unemployment.dl4 + hsassets.dl4 + CCI.dl4 +
## finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + unemployment.dl5 +
## hsassets.dl5 + CCI.dl5 + finalconsumption.dl5 + MRO.dl6 +
## OMO.dl6 + unemployment.dl6 + hsassets.dl6 + CCI.dl6 + finalconsumption.dl6 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7 8
## -4.5726 3.6679 10.9335 -30.7349 21.1006 16.9593 -36.9324 19.7214
## 9 10 11 12 13 14 15 16
## 0.2706 -19.0190 12.3349 2.2788 38.6300 -35.4453 -56.0307 75.5182
## 17 18 19 20 21 22 23 24
## -26.3348 -6.8171 4.5849 5.5529 2.2307 6.9737 18.9098 -25.9818
## 25 26 27 28 29 30 31 32
## -12.1399 7.7103 30.6677 -8.1764 -22.0606 -7.9235 6.1918 8.7717
## 33 34 35 36 37 38 39 40
## -8.3499 7.3251 -21.6149 37.5878 -13.0040 -3.9682 1.4570 -3.4957
## 41 42 43 44 45 46
## 17.7751 -16.7017 5.5181 24.9333 -5.5639 -22.7377
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 1.958e+02 5.364e+02 0.365 0.7335
## ect2 2.929e-01 5.152e-01 0.568 0.6001
## ect3 -1.861e+03 1.792e+03 -1.039 0.3577
## ect4 4.189e-04 1.213e-03 0.345 0.7472
## ect5 -7.055e+01 1.349e+02 -0.523 0.6287
## constant 2.111e+04 1.260e+04 1.675 0.1692
## MRO.dl1 -9.222e+02 3.818e+02 -2.415 0.0731 .
## OMO.dl1 -1.261e+00 4.484e-01 -2.813 0.0482 *
## unemployment.dl1 -1.797e+02 2.822e+02 -0.637 0.5590
## hsassets.dl1 2.310e-04 5.272e-04 0.438 0.6839
## CCI.dl1 2.649e-01 1.314e+01 0.020 0.9849
## finalconsumption.dl1 -1.368e-02 8.196e-03 -1.669 0.1704
## MRO.dl2 -1.318e+03 6.689e+02 -1.970 0.1202
## OMO.dl2 -8.363e-01 3.235e-01 -2.585 0.0610 .
## unemployment.dl2 -4.543e+01 4.620e+02 -0.098 0.9264
## hsassets.dl2 -2.291e-04 4.244e-04 -0.540 0.6179
## CCI.dl2 3.044e+01 3.635e+01 0.837 0.4495
## finalconsumption.dl2 -1.322e-02 7.936e-03 -1.666 0.1711
## MRO.dl3 -9.297e+02 9.084e+02 -1.023 0.3640
## OMO.dl3 1.580e-01 3.395e-01 0.465 0.6659
## unemployment.dl3 -5.477e+02 8.418e+02 -0.651 0.5507
## hsassets.dl3 6.543e-04 6.805e-04 0.962 0.3907
## CCI.dl3 5.914e+01 5.955e+01 0.993 0.3769
## finalconsumption.dl3 -2.835e-03 4.834e-03 -0.587 0.5890
## MRO.dl4 -3.096e+02 9.090e+02 -0.341 0.7506
## OMO.dl4 -5.679e-01 5.213e-01 -1.089 0.3372
## unemployment.dl4 -6.058e+02 1.290e+03 -0.470 0.6631
## hsassets.dl4 -3.335e-04 8.493e-04 -0.393 0.7146
## CCI.dl4 6.259e+01 9.422e+01 0.664 0.5429
## finalconsumption.dl4 -1.472e-02 1.308e-02 -1.125 0.3234
## MRO.dl5 -2.565e+01 8.857e+02 -0.029 0.9783
## OMO.dl5 -5.068e-01 4.823e-01 -1.051 0.3526
## unemployment.dl5 -1.601e+03 1.650e+03 -0.971 0.3866
## hsassets.dl5 -5.107e-04 8.356e-04 -0.611 0.5741
## CCI.dl5 1.338e+01 1.095e+02 0.122 0.9086
## finalconsumption.dl5 -9.442e-03 1.344e-02 -0.703 0.5210
## MRO.dl6 2.664e+02 7.228e+02 0.369 0.7311
## OMO.dl6 -4.060e-01 6.581e-01 -0.617 0.5707
## unemployment.dl6 -1.402e+03 1.830e+03 -0.766 0.4865
## hsassets.dl6 2.156e-04 1.187e-03 0.182 0.8647
## CCI.dl6 -3.148e+01 1.281e+02 -0.246 0.8180
## finalconsumption.dl6 -7.889e-03 1.606e-02 -0.491 0.6489
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 76.57 on 4 degrees of freedom
## Multiple R-squared: 0.9713, Adjusted R-squared: 0.6702
## F-statistic: 3.225 on 42 and 4 DF, p-value: 0.1304
##
##
## Response unemployment.d :
##
## Call:
## lm(formula = unemployment.d ~ ect1 + ect2 + ect3 + ect4 + ect5 +
## constant + MRO.dl1 + OMO.dl1 + unemployment.dl1 + hsassets.dl1 +
## CCI.dl1 + finalconsumption.dl1 + MRO.dl2 + OMO.dl2 + unemployment.dl2 +
## hsassets.dl2 + CCI.dl2 + finalconsumption.dl2 + MRO.dl3 +
## OMO.dl3 + unemployment.dl3 + hsassets.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + unemployment.dl4 + hsassets.dl4 + CCI.dl4 +
## finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + unemployment.dl5 +
## hsassets.dl5 + CCI.dl5 + finalconsumption.dl5 + MRO.dl6 +
## OMO.dl6 + unemployment.dl6 + hsassets.dl6 + CCI.dl6 + finalconsumption.dl6 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7 8
## 0.004657 -0.023149 0.018683 0.010569 -0.031112 0.019933 0.006941 -0.027167
## 9 10 11 12 13 14 15 16
## 0.031436 -0.020501 0.043258 -0.058672 0.024616 0.009165 0.009358 -0.040936
## 17 18 19 20 21 22 23 24
## 0.017256 -0.015313 0.049408 -0.023016 0.020407 -0.032273 -0.006363 0.002606
## 25 26 27 28 29 30 31 32
## 0.022664 -0.015385 -0.004054 0.010015 0.002109 -0.002402 -0.005270 0.013642
## 33 34 35 36 37 38 39 40
## -0.022557 0.010678 -0.014278 0.008597 0.007557 -0.007819 -0.003360 0.029433
## 41 42 43 44 45 46
## -0.031667 0.045570 -0.046048 0.009205 0.007962 -0.004382
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 8.658e-01 5.616e-01 1.542 0.1980
## ect2 -3.513e-04 5.394e-04 -0.651 0.5504
## ect3 -2.691e+00 1.876e+00 -1.434 0.2248
## ect4 1.297e-06 1.270e-06 1.022 0.3647
## ect5 -1.563e-01 1.413e-01 -1.106 0.3306
## constant 2.846e+01 1.319e+01 2.157 0.0972 .
## MRO.dl1 3.137e-01 3.998e-01 0.785 0.4765
## OMO.dl1 -4.956e-04 4.695e-04 -1.056 0.3507
## unemployment.dl1 -1.163e+00 2.955e-01 -3.935 0.0170 *
## hsassets.dl1 4.026e-07 5.520e-07 0.729 0.5062
## CCI.dl1 -2.511e-02 1.376e-02 -1.826 0.1420
## finalconsumption.dl1 -1.333e-05 8.582e-06 -1.553 0.1954
## MRO.dl2 3.131e-01 7.004e-01 0.447 0.6780
## OMO.dl2 -8.151e-04 3.387e-04 -2.406 0.0738 .
## unemployment.dl2 -1.259e+00 4.838e-01 -2.602 0.0599 .
## hsassets.dl2 -1.227e-07 4.443e-07 -0.276 0.7961
## CCI.dl2 -5.089e-02 3.806e-02 -1.337 0.2522
## finalconsumption.dl2 -1.591e-05 8.310e-06 -1.914 0.1281
## MRO.dl3 6.193e-01 9.512e-01 0.651 0.5505
## OMO.dl3 -4.270e-05 3.554e-04 -0.120 0.9102
## unemployment.dl3 -1.330e+00 8.814e-01 -1.509 0.2058
## hsassets.dl3 1.104e-06 7.125e-07 1.550 0.1962
## CCI.dl3 -6.348e-02 6.236e-02 -1.018 0.3663
## finalconsumption.dl3 -7.948e-06 5.061e-06 -1.570 0.1914
## MRO.dl4 1.188e+00 9.518e-01 1.248 0.2800
## OMO.dl4 -3.297e-04 5.458e-04 -0.604 0.5784
## unemployment.dl4 -1.884e+00 1.351e+00 -1.395 0.2355
## hsassets.dl4 1.124e-06 8.893e-07 1.264 0.2748
## CCI.dl4 -8.606e-02 9.866e-02 -0.872 0.4323
## finalconsumption.dl4 -2.044e-05 1.369e-05 -1.493 0.2097
## MRO.dl5 1.522e+00 9.274e-01 1.641 0.1761
## OMO.dl5 -3.104e-05 5.050e-04 -0.061 0.9539
## unemployment.dl5 -2.547e+00 1.727e+00 -1.475 0.2143
## hsassets.dl5 2.720e-07 8.749e-07 0.311 0.7714
## CCI.dl5 -1.105e-01 1.147e-01 -0.963 0.3900
## finalconsumption.dl5 -1.836e-05 1.407e-05 -1.305 0.2620
## MRO.dl6 1.240e+00 7.569e-01 1.639 0.1766
## OMO.dl6 -8.234e-04 6.891e-04 -1.195 0.2981
## unemployment.dl6 -2.897e+00 1.917e+00 -1.511 0.2052
## hsassets.dl6 2.599e-07 1.243e-06 0.209 0.8446
## CCI.dl6 -1.575e-01 1.342e-01 -1.174 0.3054
## finalconsumption.dl6 -2.125e-05 1.681e-05 -1.264 0.2748
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.08017 on 4 degrees of freedom
## Multiple R-squared: 0.9856, Adjusted R-squared: 0.8348
## F-statistic: 6.536 on 42 and 4 DF, p-value: 0.0394
##
##
## Response hsassets.d :
##
## Call:
## lm(formula = hsassets.d ~ ect1 + ect2 + ect3 + ect4 + ect5 +
## constant + MRO.dl1 + OMO.dl1 + unemployment.dl1 + hsassets.dl1 +
## CCI.dl1 + finalconsumption.dl1 + MRO.dl2 + OMO.dl2 + unemployment.dl2 +
## hsassets.dl2 + CCI.dl2 + finalconsumption.dl2 + MRO.dl3 +
## OMO.dl3 + unemployment.dl3 + hsassets.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + unemployment.dl4 + hsassets.dl4 + CCI.dl4 +
## finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + unemployment.dl5 +
## hsassets.dl5 + CCI.dl5 + finalconsumption.dl5 + MRO.dl6 +
## OMO.dl6 + unemployment.dl6 + hsassets.dl6 + CCI.dl6 + finalconsumption.dl6 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7 8
## 835.3 -6429.4 7701.1 9636.8 -26878.8 16059.7 14199.0 -30141.7
## 9 10 11 12 13 14 15 16
## 27920.1 -5402.0 -17031.5 -1075.8 -4555.2 44678.4 -24886.3 -37313.0
## 17 18 19 20 21 22 23 24
## 46865.4 -7249.8 -13506.0 11086.0 -11534.7 4973.8 -3629.5 22136.5
## 25 26 27 28 29 30 31 32
## -10854.4 -13097.8 12207.1 -5994.3 -5276.3 8624.1 -1225.7 3903.5
## 33 34 35 36 37 38 39 40
## -5374.1 4904.8 -7289.5 6464.5 1673.5 -20450.8 7885.2 5673.0
## 41 42 43 44 45 46
## 5158.2 -5788.6 -3503.4 1132.7 13480.8 -8711.1
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 -1.084e+06 3.872e+05 -2.800 0.04880 *
## ect2 6.382e+02 3.718e+02 1.716 0.16124
## ect3 5.112e+06 1.293e+06 3.954 0.01677 *
## ect4 -3.096e+00 8.752e-01 -3.537 0.02407 *
## ect5 3.563e+05 9.738e+04 3.659 0.02161 *
## constant -2.945e+07 9.096e+06 -3.237 0.03175 *
## MRO.dl1 -1.703e+03 2.756e+05 -0.006 0.99537
## OMO.dl1 2.122e+02 3.236e+02 0.656 0.54788
## unemployment.dl1 4.776e+05 2.037e+05 2.344 0.07901 .
## hsassets.dl1 -6.516e-01 3.805e-01 -1.713 0.16196
## CCI.dl1 2.862e+04 9.484e+03 3.018 0.03923 *
## finalconsumption.dl1 2.649e+01 5.916e+00 4.478 0.01101 *
## MRO.dl2 -6.196e+05 4.828e+05 -1.283 0.26871
## OMO.dl2 -5.561e+02 2.335e+02 -2.382 0.07586 .
## unemployment.dl2 1.407e+06 3.335e+05 4.218 0.01350 *
## hsassets.dl2 -1.023e+00 3.063e-01 -3.341 0.02882 *
## CCI.dl2 6.987e+04 2.624e+04 2.663 0.05620 .
## finalconsumption.dl2 1.618e+01 5.728e+00 2.825 0.04757 *
## MRO.dl3 -1.499e+06 6.557e+05 -2.287 0.08416 .
## OMO.dl3 2.325e+02 2.450e+02 0.949 0.39640
## unemployment.dl3 2.487e+06 6.076e+05 4.093 0.01494 *
## hsassets.dl3 -1.885e+00 4.911e-01 -3.839 0.01848 *
## CCI.dl3 1.192e+05 4.299e+04 2.773 0.05019 .
## finalconsumption.dl3 1.698e+01 3.489e+00 4.866 0.00824 **
## MRO.dl4 -1.911e+06 6.561e+05 -2.913 0.04357 *
## OMO.dl4 6.887e+02 3.762e+02 1.831 0.14114
## unemployment.dl4 3.938e+06 9.311e+05 4.230 0.01337 *
## hsassets.dl4 -1.875e+00 6.130e-01 -3.058 0.03773 *
## CCI.dl4 1.980e+05 6.801e+04 2.912 0.04361 *
## finalconsumption.dl4 3.483e+01 9.438e+00 3.690 0.02101 *
## MRO.dl5 -2.067e+06 6.393e+05 -3.233 0.03189 *
## OMO.dl5 1.098e+03 3.481e+02 3.153 0.03442 *
## unemployment.dl5 4.501e+06 1.191e+06 3.780 0.01943 *
## hsassets.dl5 -2.239e+00 6.031e-01 -3.712 0.02062 *
## CCI.dl5 2.759e+05 7.905e+04 3.490 0.02512 *
## finalconsumption.dl5 3.578e+01 9.698e+00 3.689 0.02104 *
## MRO.dl6 -1.540e+06 5.217e+05 -2.952 0.04188 *
## OMO.dl6 1.136e+03 4.750e+02 2.392 0.07497 .
## unemployment.dl6 5.112e+06 1.321e+06 3.869 0.01801 *
## hsassets.dl6 -3.231e+00 8.569e-01 -3.770 0.01960 *
## CCI.dl6 3.324e+05 9.249e+04 3.594 0.02289 *
## finalconsumption.dl6 4.220e+01 1.159e+01 3.641 0.02194 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 55260 on 4 degrees of freedom
## Multiple R-squared: 0.9945, Adjusted R-squared: 0.9364
## F-statistic: 17.12 on 42 and 4 DF, p-value: 0.006566
##
##
## Response CCI.d :
##
## Call:
## lm(formula = CCI.d ~ ect1 + ect2 + ect3 + ect4 + ect5 + constant +
## MRO.dl1 + OMO.dl1 + unemployment.dl1 + hsassets.dl1 + CCI.dl1 +
## finalconsumption.dl1 + MRO.dl2 + OMO.dl2 + unemployment.dl2 +
## hsassets.dl2 + CCI.dl2 + finalconsumption.dl2 + MRO.dl3 +
## OMO.dl3 + unemployment.dl3 + hsassets.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + unemployment.dl4 + hsassets.dl4 + CCI.dl4 +
## finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + unemployment.dl5 +
## hsassets.dl5 + CCI.dl5 + finalconsumption.dl5 + MRO.dl6 +
## OMO.dl6 + unemployment.dl6 + hsassets.dl6 + CCI.dl6 + finalconsumption.dl6 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7 8
## 0.07936 -0.04548 -0.24809 0.18988 0.36243 -0.61892 0.07915 0.53629
## 9 10 11 12 13 14 15 16
## -0.69667 0.26210 0.85181 -0.55975 -0.14260 -1.00779 1.76534 -0.12450
## 17 18 19 20 21 22 23 24
## -1.13573 0.20668 0.89422 -0.69564 0.57713 -0.58683 -0.19753 -0.39971
## 25 26 27 28 29 30 31 32
## 0.77650 0.21058 -0.90548 0.42537 0.51851 -0.21958 -0.09409 -0.13086
## 33 34 35 36 37 38 39 40
## 0.09140 -0.17487 0.42596 -0.67571 0.19588 0.70999 -0.33424 0.13069
## 41 42 43 44 45 46
## -0.73930 0.88086 -0.39683 -0.30185 -0.32601 0.58794
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 1.884e+01 1.380e+01 1.365 0.2440
## ect2 3.940e-02 1.326e-02 2.972 0.0411 *
## ect3 6.269e-01 4.611e+01 0.014 0.9898
## ect4 -1.100e-05 3.120e-05 -0.353 0.7422
## ect5 -6.867e-01 3.472e+00 -0.198 0.8529
## constant 2.747e+02 3.243e+02 0.847 0.4447
## MRO.dl1 2.950e+00 9.826e+00 0.300 0.7789
## OMO.dl1 -5.121e-03 1.154e-02 -0.444 0.6801
## unemployment.dl1 -1.140e+00 7.263e+00 -0.157 0.8829
## hsassets.dl1 -1.366e-06 1.357e-05 -0.101 0.9247
## CCI.dl1 -1.071e+00 3.381e-01 -3.166 0.0340 *
## finalconsumption.dl1 1.062e-04 2.109e-04 0.503 0.6411
## MRO.dl2 5.268e+00 1.721e+01 0.306 0.7749
## OMO.dl2 -3.012e-03 8.325e-03 -0.362 0.7358
## unemployment.dl2 -1.510e+01 1.189e+01 -1.270 0.2729
## hsassets.dl2 2.338e-05 1.092e-05 2.141 0.0990 .
## CCI.dl2 -1.352e+00 9.355e-01 -1.445 0.2219
## finalconsumption.dl2 -5.632e-07 2.042e-04 -0.003 0.9979
## MRO.dl3 -4.484e+00 2.338e+01 -0.192 0.8572
## OMO.dl3 -6.643e-03 8.736e-03 -0.760 0.4894
## unemployment.dl3 -1.073e+01 2.166e+01 -0.495 0.6463
## hsassets.dl3 -9.949e-06 1.751e-05 -0.568 0.6003
## CCI.dl3 -1.614e+00 1.533e+00 -1.053 0.3517
## finalconsumption.dl3 -1.571e-04 1.244e-04 -1.263 0.2752
## MRO.dl4 -9.583e+00 2.339e+01 -0.410 0.7030
## OMO.dl4 1.343e-02 1.341e-02 1.001 0.3735
## unemployment.dl4 -6.864e+00 3.320e+01 -0.207 0.8463
## hsassets.dl4 1.404e-06 2.186e-05 0.064 0.9519
## CCI.dl4 -9.773e-01 2.425e+00 -0.403 0.7075
## finalconsumption.dl4 -8.648e-05 3.365e-04 -0.257 0.8099
## MRO.dl5 -1.602e+01 2.279e+01 -0.703 0.5208
## OMO.dl5 1.396e-02 1.241e-02 1.125 0.3236
## unemployment.dl5 1.961e+00 4.245e+01 0.046 0.9654
## hsassets.dl5 6.614e-06 2.150e-05 0.308 0.7738
## CCI.dl5 -8.820e-01 2.819e+00 -0.313 0.7700
## finalconsumption.dl5 -1.921e-04 3.458e-04 -0.556 0.6081
## MRO.dl6 8.112e-01 1.860e+01 0.044 0.9673
## OMO.dl6 4.174e-02 1.694e-02 2.465 0.0694 .
## unemployment.dl6 -1.503e+00 4.711e+01 -0.032 0.9761
## hsassets.dl6 -1.305e-05 3.055e-05 -0.427 0.6912
## CCI.dl6 -6.192e-02 3.298e+00 -0.019 0.9859
## finalconsumption.dl6 -1.015e-04 4.132e-04 -0.246 0.8181
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.97 on 4 degrees of freedom
## Multiple R-squared: 0.9567, Adjusted R-squared: 0.502
## F-statistic: 2.104 on 42 and 4 DF, p-value: 0.2464
##
##
## Response finalconsumption.d :
##
## Call:
## lm(formula = finalconsumption.d ~ ect1 + ect2 + ect3 + ect4 +
## ect5 + constant + MRO.dl1 + OMO.dl1 + unemployment.dl1 +
## hsassets.dl1 + CCI.dl1 + finalconsumption.dl1 + MRO.dl2 +
## OMO.dl2 + unemployment.dl2 + hsassets.dl2 + CCI.dl2 + finalconsumption.dl2 +
## MRO.dl3 + OMO.dl3 + unemployment.dl3 + hsassets.dl3 + CCI.dl3 +
## finalconsumption.dl3 + MRO.dl4 + OMO.dl4 + unemployment.dl4 +
## hsassets.dl4 + CCI.dl4 + finalconsumption.dl4 + MRO.dl5 +
## OMO.dl5 + unemployment.dl5 + hsassets.dl5 + CCI.dl5 + finalconsumption.dl5 +
## MRO.dl6 + OMO.dl6 + unemployment.dl6 + hsassets.dl6 + CCI.dl6 +
## finalconsumption.dl6 - 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7 8
## -132.94 502.87 -248.46 34.28 -190.67 136.50 373.40 -494.37
## 9 10 11 12 13 14 15 16
## 281.39 407.28 -2335.28 2012.99 -999.50 1810.74 -1695.45 -234.77
## 17 18 19 20 21 22 23 24
## 1656.51 162.19 -2394.70 1371.52 -1283.77 1396.84 88.10 945.23
## 25 26 27 28 29 30 31 32
## -1363.20 -91.11 820.32 -668.44 -387.89 493.49 138.91 -264.88
## 33 34 35 36 37 38 39 40
## 506.05 -114.49 88.93 107.38 -213.13 -749.18 512.63 -760.73
## 41 42 43 44 45 46
## 1410.20 -1927.30 1456.75 -208.32 375.90 -331.84
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 -3.375e+04 2.374e+04 -1.421 0.2283
## ect2 -3.163e+01 2.280e+01 -1.387 0.2377
## ect3 1.159e+05 7.930e+04 1.461 0.2178
## ect4 -2.568e-02 5.367e-02 -0.478 0.6573
## ect5 5.715e+03 5.972e+03 0.957 0.3928
## constant -1.443e+06 5.578e+05 -2.587 0.0609 .
## MRO.dl1 2.451e+04 1.690e+04 1.450 0.2206
## OMO.dl1 3.849e+01 1.985e+01 1.939 0.1245
## unemployment.dl1 1.430e+04 1.249e+04 1.145 0.3161
## hsassets.dl1 -1.754e-02 2.333e-02 -0.752 0.4941
## CCI.dl1 2.062e+02 5.816e+02 0.354 0.7409
## finalconsumption.dl1 8.166e-01 3.628e-01 2.251 0.0876 .
## MRO.dl2 4.791e+04 2.961e+04 1.618 0.1810
## OMO.dl2 2.526e+01 1.432e+01 1.764 0.1525
## unemployment.dl2 2.162e+04 2.045e+04 1.057 0.3501
## hsassets.dl2 1.165e-02 1.878e-02 0.620 0.5686
## CCI.dl2 3.310e+02 1.609e+03 0.206 0.8471
## finalconsumption.dl2 8.086e-01 3.513e-01 2.302 0.0828 .
## MRO.dl3 2.520e+04 4.021e+04 0.627 0.5649
## OMO.dl3 -2.606e+01 1.503e+01 -1.734 0.1579
## unemployment.dl3 3.808e+04 3.726e+04 1.022 0.3645
## hsassets.dl3 -2.651e-02 3.012e-02 -0.880 0.4284
## CCI.dl3 -3.932e+02 2.636e+03 -0.149 0.8887
## finalconsumption.dl3 -3.959e-01 2.139e-01 -1.850 0.1379
## MRO.dl4 -3.414e+03 4.024e+04 -0.085 0.9365
## OMO.dl4 -2.046e+00 2.307e+01 -0.089 0.9336
## unemployment.dl4 5.065e+04 5.710e+04 0.887 0.4252
## hsassets.dl4 -2.618e-02 3.759e-02 -0.696 0.5246
## CCI.dl4 2.983e+01 4.171e+03 0.007 0.9946
## finalconsumption.dl4 8.864e-01 5.788e-01 1.531 0.2004
## MRO.dl5 -1.971e+04 3.920e+04 -0.503 0.6415
## OMO.dl5 -9.934e+00 2.135e+01 -0.465 0.6659
## unemployment.dl5 7.937e+04 7.302e+04 1.087 0.3382
## hsassets.dl5 4.471e-03 3.699e-02 0.121 0.9096
## CCI.dl5 1.501e+03 4.848e+03 0.310 0.7723
## finalconsumption.dl5 6.396e-01 5.948e-01 1.075 0.3428
## MRO.dl6 -3.082e+04 3.200e+04 -0.963 0.3899
## OMO.dl6 1.985e+00 2.913e+01 0.068 0.9489
## unemployment.dl6 1.111e+05 8.102e+04 1.372 0.2420
## hsassets.dl6 -1.414e-02 5.255e-02 -0.269 0.8011
## CCI.dl6 3.915e+03 5.672e+03 0.690 0.5280
## finalconsumption.dl6 2.298e-01 7.107e-01 0.323 0.7626
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3389 on 4 degrees of freedom
## Multiple R-squared: 0.9971, Adjusted R-squared: 0.9667
## F-statistic: 32.81 on 42 and 4 DF, p-value: 0.001862## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 2 2 2 2
##
## $criteria
## 1 2 3 4 5 6 7 8 9 10
## AIC(n) 1.725233e+01 -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## HQ(n) 1.745695e+01 -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## SC(n) 1.931086e+01 -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## FPE(n) 4.245549e+07 0 0 0 0 0 0 0 0 0##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 1.000000000 0.984585598 0.853794918 0.601643749 0.369522689 0.002160951
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 5 | 0.05 6.50 8.18 11.65
## r <= 4 | 11.12 15.66 17.95 23.52
## r <= 3 | 33.21 28.71 31.52 37.22
## r <= 2 | 79.36 45.23 48.28 55.43
## r <= 1 | 179.50 66.49 70.60 78.87
## r = 0 | 784.70 85.18 90.39 104.20
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## MRO.l3 OMO.l3 unemployment.l3 hsassets.l3
## MRO.l3 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## OMO.l3 5.813504e-03 1.382640e-01 1.498658e-03 -2.026272e-02
## unemployment.l3 -3.214286e+00 3.120267e+02 1.066859e+00 -4.510428e+00
## hsassets.l3 1.521976e-06 -2.519329e-05 1.935560e-08 -1.161885e-05
## CCI.l3 -1.519472e-01 -2.028437e+01 -3.201207e-03 -3.850752e+00
## durablehs.l3 1.009241e+00 6.946468e+01 -1.619949e-01 3.332629e+00
## CCI.l3 durablehs.l3
## MRO.l3 1.000000e+00 1.0000000000
## OMO.l3 9.521055e-02 0.0069303569
## unemployment.l3 -5.301641e+01 0.6647163244
## hsassets.l3 4.015034e-05 -0.0000011502
## CCI.l3 -6.893355e+00 0.4508507752
## durablehs.l3 1.159872e+01 0.4702218487
##
## Weights W:
## (This is the loading matrix)
##
## MRO.l3 OMO.l3 unemployment.l3 hsassets.l3
## MRO.d -1.747550e-01 -0.002168977 3.566086e-01 3.574325e-03
## OMO.d -5.227125e+01 1.569858227 -5.647864e+02 -1.808129e-01
## unemployment.d 4.716436e-02 -0.001763833 -3.941686e-01 2.124384e-02
## hsassets.d -2.030031e+04 306.311605642 3.368229e+05 1.482918e+04
## CCI.d 6.755020e-01 0.003374714 4.381478e+00 -5.166128e-02
## durablehs.d -3.035797e-01 -0.005955248 6.932723e-01 2.036183e-03
## CCI.l3 durablehs.l3
## MRO.d 4.350792e-03 -1.332444e-03
## OMO.d -1.481671e+00 1.513096e+00
## unemployment.d 1.554368e-03 9.438740e-04
## hsassets.d -1.507389e+03 -4.417717e+02
## CCI.d 1.346434e-02 1.897667e-02
## durablehs.d -1.786835e-03 1.371174e-03## Response MRO.d :
##
## Call:
## lm(formula = MRO.d ~ ect1 + ect2 + ect3 + ect4 + constant + MRO.dl1 +
## OMO.dl1 + unemployment.dl1 + hsassets.dl1 + CCI.dl1 + durablehs.dl1 +
## MRO.dl2 + OMO.dl2 + unemployment.dl2 + hsassets.dl2 + CCI.dl2 +
## durablehs.dl2 - 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.19686 -0.04342 0.01226 0.03562 0.21941
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 1.833e-01 2.257e-01 0.812 0.44358
## ect2 -8.538e-04 4.679e-04 -1.825 0.11078
## ect3 2.493e-01 4.235e-01 0.589 0.57466
## ect4 -2.460e-07 1.177e-07 -2.090 0.07499 .
## constant 8.053e+00 4.582e+00 1.757 0.12225
## MRO.dl1 -1.813e-01 3.065e-01 -0.592 0.57268
## OMO.dl1 5.275e-04 4.881e-04 1.081 0.31568
## unemployment.dl1 3.229e-01 3.063e-01 1.054 0.32675
## hsassets.dl1 -4.572e-07 5.495e-07 -0.832 0.43281
## CCI.dl1 1.171e-01 3.206e-02 3.653 0.00815 **
## durablehs.dl1 -4.605e-01 1.978e-01 -2.328 0.05274 .
## MRO.dl2 -1.045e-01 3.186e-01 -0.328 0.75260
## OMO.dl2 1.606e-04 4.991e-04 0.322 0.75697
## unemployment.dl2 -2.860e-01 2.836e-01 -1.008 0.34687
## hsassets.dl2 -2.486e-07 3.746e-07 -0.664 0.52815
## CCI.dl2 8.031e-03 3.869e-02 0.208 0.84147
## durablehs.dl2 -3.371e-01 1.537e-01 -2.193 0.06442 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1486 on 7 degrees of freedom
## Multiple R-squared: 0.9349, Adjusted R-squared: 0.7769
## F-statistic: 5.915 on 17 and 7 DF, p-value: 0.01169
##
##
## Response OMO.d :
##
## Call:
## lm(formula = OMO.d ~ ect1 + ect2 + ect3 + ect4 + constant + MRO.dl1 +
## OMO.dl1 + unemployment.dl1 + hsassets.dl1 + CCI.dl1 + durablehs.dl1 +
## MRO.dl2 + OMO.dl2 + unemployment.dl2 + hsassets.dl2 + CCI.dl2 +
## durablehs.dl2 - 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -83.657 -23.106 -9.704 15.521 152.190
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 -6.157e+02 1.313e+02 -4.690 0.00223 **
## ect2 -9.296e-01 2.721e-01 -3.416 0.01119 *
## ect3 5.612e+01 2.463e+02 0.228 0.82628
## ect4 -1.279e-04 6.845e-05 -1.869 0.10381
## constant 5.794e+03 2.665e+03 2.174 0.06619 .
## MRO.dl1 -6.026e+02 1.782e+02 -3.381 0.01175 *
## OMO.dl1 -1.286e+00 2.839e-01 -4.531 0.00270 **
## unemployment.dl1 3.825e+01 1.781e+02 0.215 0.83609
## hsassets.dl1 -9.879e-05 3.196e-04 -0.309 0.76621
## CCI.dl1 -8.160e+01 1.865e+01 -4.376 0.00325 **
## durablehs.dl1 1.805e+02 1.150e+02 1.569 0.16065
## MRO.dl2 -5.502e+02 1.853e+02 -2.969 0.02083 *
## OMO.dl2 -1.130e+00 2.903e-01 -3.893 0.00595 **
## unemployment.dl2 -2.858e+01 1.649e+02 -0.173 0.86733
## hsassets.dl2 1.940e-04 2.179e-04 0.891 0.40268
## CCI.dl2 -1.854e+01 2.250e+01 -0.824 0.43706
## durablehs.dl2 3.439e+02 8.941e+01 3.846 0.00633 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 86.42 on 7 degrees of freedom
## Multiple R-squared: 0.9231, Adjusted R-squared: 0.7363
## F-statistic: 4.941 on 17 and 7 DF, p-value: 0.01954
##
##
## Response unemployment.d :
##
## Call:
## lm(formula = unemployment.d ~ ect1 + ect2 + ect3 + ect4 + constant +
## MRO.dl1 + OMO.dl1 + unemployment.dl1 + hsassets.dl1 + CCI.dl1 +
## durablehs.dl1 + MRO.dl2 + OMO.dl2 + unemployment.dl2 + hsassets.dl2 +
## CCI.dl2 + durablehs.dl2 - 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.165030 -0.049426 0.006364 0.037386 0.154051
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 -3.275e-01 2.098e-01 -1.561 0.16254
## ect2 -9.909e-04 4.350e-04 -2.278 0.05681 .
## ect3 -1.218e+00 3.937e-01 -3.094 0.01746 *
## ect4 -1.382e-07 1.094e-07 -1.263 0.24688
## constant 1.316e+01 4.260e+00 3.089 0.01758 *
## MRO.dl1 2.081e-01 2.849e-01 0.730 0.48893
## OMO.dl1 -1.171e-04 4.538e-04 -0.258 0.80380
## unemployment.dl1 -4.570e-01 2.847e-01 -1.605 0.15249
## hsassets.dl1 -6.424e-09 5.108e-07 -0.013 0.99032
## CCI.dl1 4.956e-03 2.981e-02 0.166 0.87266
## durablehs.dl1 5.280e-02 1.839e-01 0.287 0.78228
## MRO.dl2 3.048e-01 2.962e-01 1.029 0.33776
## OMO.dl2 -4.292e-04 4.640e-04 -0.925 0.38570
## unemployment.dl2 -9.755e-01 2.636e-01 -3.700 0.00766 **
## hsassets.dl2 6.699e-08 3.482e-07 0.192 0.85293
## CCI.dl2 2.749e-02 3.597e-02 0.764 0.46968
## durablehs.dl2 -8.434e-02 1.429e-01 -0.590 0.57365
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1381 on 7 degrees of freedom
## Multiple R-squared: 0.8265, Adjusted R-squared: 0.4052
## F-statistic: 1.962 on 17 and 7 DF, p-value: 0.1852
##
##
## Response hsassets.d :
##
## Call:
## lm(formula = hsassets.d ~ ect1 + ect2 + ect3 + ect4 + constant +
## MRO.dl1 + OMO.dl1 + unemployment.dl1 + hsassets.dl1 + CCI.dl1 +
## durablehs.dl1 + MRO.dl2 + OMO.dl2 + unemployment.dl2 + hsassets.dl2 +
## CCI.dl2 + durablehs.dl2 - 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -123920 -29331 4887 36431 133536
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 3.317e+05 1.561e+05 2.125 0.0712 .
## ect2 1.286e+02 3.236e+02 0.398 0.7028
## ect3 4.533e+05 2.929e+05 1.548 0.1656
## ect4 -2.044e-01 8.139e-02 -2.511 0.0403 *
## constant 1.132e+06 3.169e+06 0.357 0.7314
## MRO.dl1 -2.325e+05 2.119e+05 -1.097 0.3089
## OMO.dl1 -4.978e+01 3.376e+02 -0.147 0.8869
## unemployment.dl1 -8.037e+04 2.118e+05 -0.379 0.7156
## hsassets.dl1 -1.725e-01 3.800e-01 -0.454 0.6636
## CCI.dl1 -6.879e+04 2.217e+04 -3.102 0.0173 *
## durablehs.dl1 -1.506e+04 1.368e+05 -0.110 0.9154
## MRO.dl2 4.870e+04 2.204e+05 0.221 0.8314
## OMO.dl2 1.951e+01 3.452e+02 0.057 0.9565
## unemployment.dl2 3.922e+05 1.961e+05 2.000 0.0856 .
## hsassets.dl2 1.050e-02 2.591e-01 0.041 0.9688
## CCI.dl2 -4.675e+04 2.676e+04 -1.747 0.1241
## durablehs.dl2 -4.128e+04 1.063e+05 -0.388 0.7093
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 102800 on 7 degrees of freedom
## Multiple R-squared: 0.936, Adjusted R-squared: 0.7806
## F-statistic: 6.024 on 17 and 7 DF, p-value: 0.01109
##
##
## Response CCI.d :
##
## Call:
## lm(formula = CCI.d ~ ect1 + ect2 + ect3 + ect4 + constant + MRO.dl1 +
## OMO.dl1 + unemployment.dl1 + hsassets.dl1 + CCI.dl1 + durablehs.dl1 +
## MRO.dl2 + OMO.dl2 + unemployment.dl2 + hsassets.dl2 + CCI.dl2 +
## durablehs.dl2 - 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.71810 -0.35493 -0.04002 0.33853 1.14080
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 5.009e+00 1.304e+00 3.840 0.00638 **
## ect2 1.201e-02 2.704e-03 4.441 0.00301 **
## ect3 3.789e+00 2.448e+00 1.548 0.16551
## ect4 1.628e-06 6.801e-07 2.394 0.04789 *
## constant -1.043e+02 2.648e+01 -3.939 0.00561 **
## MRO.dl1 5.336e+00 1.771e+00 3.013 0.01957 *
## OMO.dl1 6.444e-03 2.821e-03 2.285 0.05626 .
## unemployment.dl1 3.302e-01 1.770e+00 0.187 0.85728
## hsassets.dl1 3.786e-06 3.175e-06 1.192 0.27195
## CCI.dl1 -2.396e-01 1.853e-01 -1.293 0.23709
## durablehs.dl1 9.181e-01 1.143e+00 0.803 0.44821
## MRO.dl2 5.859e+00 1.841e+00 3.182 0.01545 *
## OMO.dl2 6.064e-03 2.884e-03 2.103 0.07359 .
## unemployment.dl2 1.415e+00 1.639e+00 0.864 0.41638
## hsassets.dl2 3.791e-06 2.165e-06 1.751 0.12336
## CCI.dl2 -2.803e-01 2.236e-01 -1.254 0.25018
## durablehs.dl2 -2.294e+00 8.884e-01 -2.583 0.03633 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8587 on 7 degrees of freedom
## Multiple R-squared: 0.941, Adjusted R-squared: 0.7977
## F-statistic: 6.566 on 17 and 7 DF, p-value: 0.008606
##
##
## Response durablehs.d :
##
## Call:
## lm(formula = durablehs.d ~ ect1 + ect2 + ect3 + ect4 + constant +
## MRO.dl1 + OMO.dl1 + unemployment.dl1 + hsassets.dl1 + CCI.dl1 +
## durablehs.dl1 + MRO.dl2 + OMO.dl2 + unemployment.dl2 + hsassets.dl2 +
## CCI.dl2 + durablehs.dl2 - 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.110033 -0.045890 -0.004932 0.048166 0.095473
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 3.858e-01 1.636e-01 2.358 0.05050 .
## ect2 -1.591e-03 3.392e-04 -4.690 0.00224 **
## ect3 -1.520e-01 3.070e-01 -0.495 0.63576
## ect4 -3.223e-07 8.531e-08 -3.777 0.00692 **
## constant 1.535e+01 3.321e+00 4.620 0.00243 **
## MRO.dl1 -5.105e-01 2.221e-01 -2.298 0.05515 .
## OMO.dl1 -4.971e-04 3.538e-04 -1.405 0.20286
## unemployment.dl1 -6.976e-01 2.220e-01 -3.142 0.01633 *
## hsassets.dl1 8.249e-07 3.983e-07 2.071 0.07711 .
## CCI.dl1 -8.554e-03 2.324e-02 -0.368 0.72371
## durablehs.dl1 -7.420e-01 1.434e-01 -5.176 0.00129 **
## MRO.dl2 8.364e-02 2.310e-01 0.362 0.72792
## OMO.dl2 -9.798e-04 3.618e-04 -2.708 0.03027 *
## unemployment.dl2 -8.075e-02 2.056e-01 -0.393 0.70613
## hsassets.dl2 1.894e-07 2.715e-07 0.697 0.50804
## CCI.dl2 1.443e-01 2.804e-02 5.145 0.00133 **
## durablehs.dl2 -7.809e-01 1.114e-01 -7.008 0.00021 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1077 on 7 degrees of freedom
## Multiple R-squared: 0.9919, Adjusted R-squared: 0.9723
## F-statistic: 50.49 on 17 and 7 DF, p-value: 1.11e-05## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 6 6 6 7
##
## $criteria
## 1 2 3 4 5 6 7 8
## AIC(n) 8.925111 7.270705 5.530586 3.282407 -7.80541952 -Inf -Inf -Inf
## HQ(n) 9.770941 8.856635 7.856617 6.348538 -3.99918706 -Inf -Inf -Inf
## SC(n) 11.218767 11.571310 11.838140 11.596910 2.51603233 -Inf -Inf -Inf
## FPE(n) 7752.753211 1780.507796 532.109247 201.101109 0.07718225 NaN 0 0
## 9 10
## AIC(n) -Inf -Inf
## HQ(n) -Inf -Inf
## SC(n) -Inf -Inf
## FPE(n) 0 0##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 7.654872e+05 1.000000e+00 9.999999e-01 9.660297e-01 7.855361e-01
## [6] 7.025982e-01 2.397441e-01
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 6 | 12.88 6.50 8.18 11.65
## r <= 5 | 69.88 15.66 17.95 23.52
## r <= 4 | 142.24 28.71 31.52 37.22
## r <= 3 | 301.21 45.23 48.28 55.43
## r <= 2 | 1084.06 66.49 70.60 78.87
## r <= 1 | NaN 85.18 90.39 104.20
## r = 0 | NaN 118.99 124.25 136.06
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## MRO.l6 OMO.l6 GDP.l6
## MRO.l6 1.000000e+00 1.000000e+00 1.000000e+00
## OMO.l6 2.100636e-03 -2.878012e-03 3.360359e-03
## GDP.l6 -4.395066e+01 -1.232124e-07 2.079349e-06
## unemployment.l6 7.278035e-01 7.114716e-01 -4.569070e+00
## loanshsannualgrowth.l6 1.913711e+00 1.838722e+00 -1.084856e+00
## CCI.l6 -6.301577e-02 4.820420e-02 -1.120589e-01
## finalconsumption.l6 2.354658e-06 -2.519107e-06 -2.517958e-05
## unemployment.l6 loanshsannualgrowth.l6 CCI.l6
## MRO.l6 1.000000e+00 1.0000000000 1.000000e+00
## OMO.l6 7.373140e-03 0.0036439574 5.589557e-03
## GDP.l6 -8.138147e+00 -1.3570740684 -6.333572e+00
## unemployment.l6 -1.153390e+01 -4.1201501871 -4.434389e+00
## loanshsannualgrowth.l6 -4.456692e+00 -1.4634037903 -1.561496e+00
## CCI.l6 -1.791637e-01 -0.0808928187 -6.041236e-02
## finalconsumption.l6 -6.604416e-05 -0.0000244023 -2.119001e-05
## finalconsumption.l6
## MRO.l6 1.000000e+00
## OMO.l6 1.043314e-04
## GDP.l6 -6.424780e+00
## unemployment.l6 1.446468e+00
## loanshsannualgrowth.l6 1.371781e+00
## CCI.l6 -7.475148e-03
## finalconsumption.l6 3.285635e-06
##
## Weights W:
## (This is the loading matrix)
##
## MRO.l6 OMO.l6 GDP.l6 unemployment.l6
## MRO.d 5.623619e-01 -1.989026e-01 7.532306e-01 -0.23961500
## OMO.d -1.390669e+00 -1.187896e+02 -2.016401e+01 233.81158977
## GDP.d 1.026297e-02 5.158594e-03 -5.050059e-02 0.01032482
## unemployment.d 7.374073e-01 -6.493780e-02 3.461231e-02 0.05405793
## loanshsannualgrowth.d -1.584697e-02 -2.184747e-01 -2.343017e-01 0.05526582
## CCI.d 1.704855e+01 -2.028079e+00 4.454542e+00 0.87895088
## finalconsumption.d -9.095201e+03 1.774696e+04 2.500487e+04 -165.88390130
## loanshsannualgrowth.l6 CCI.l6 finalconsumption.l6
## MRO.d -0.46791294 0.17104695 -3.588450e-01
## OMO.d -55.66530498 -211.78850486 3.016548e+02
## GDP.d -0.04492638 0.01108342 -2.532564e-02
## unemployment.d 0.15622740 -0.26223021 -8.891129e-01
## loanshsannualgrowth.d -0.11267841 0.15926493 5.848811e-02
## CCI.d 13.83981248 -1.86454124 2.734992e+00
## finalconsumption.d 9837.32074801 8196.99106448 2.052009e+04## Response MRO.d :
##
## Call:
## lm(formula = MRO.d ~ ect1 + ect2 + ect3 + ect4 + ect5 + ect6 +
## constant + MRO.dl1 + OMO.dl1 + GDP.dl1 + unemployment.dl1 +
## loanshsannualgrowth.dl1 + CCI.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + GDP.dl2 + unemployment.dl2 + loanshsannualgrowth.dl2 +
## CCI.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 + GDP.dl3 +
## unemployment.dl3 + loanshsannualgrowth.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + GDP.dl4 + unemployment.dl4 + loanshsannualgrowth.dl4 +
## CCI.dl4 + finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + GDP.dl5 +
## unemployment.dl5 + loanshsannualgrowth.dl5 + CCI.dl5 + finalconsumption.dl5 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.0006958 0.0076016 -0.0086864 0.0004695 0.0055816 0.0049366 -0.0160179
## 8 9 10 11 12 13 14
## 0.0141768 -0.0044953 0.0044844 -0.0173373 0.0104212 -0.0407265 0.0138735
## 15 16 17 18 19 20 21
## 0.0094217 0.0366599 -0.0080771 -0.0726944 0.0290744 -0.0219480 0.0504751
## 22 23 24 25 26 27 28
## 0.0097289 -0.0143565 -0.0099596 -0.0010490 0.0090404 0.0082764 -0.0339880
## 29 30 31 32 33 34 35
## 0.0112047 0.0300112 -0.0043443 -0.0062832 -0.0083747 0.0015510 -0.0052183
## 36 37 38 39 40 41 42
## 0.0133028 -0.0037334 -0.0194923 0.0126063 -0.0059901 0.0119828 -0.0077245
## 43 44 45 46 47
## -0.0053848 0.0104541 -0.0155766 -0.0305076 0.0573269
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 5.278e-01 3.372e-01 1.565 0.178305
## ect2 1.659e-03 1.174e-03 1.414 0.216590
## ect3 -2.091e+01 9.055e+00 -2.309 0.068992 .
## ect4 7.211e-01 1.284e+00 0.562 0.598590
## ect5 1.278e+00 6.118e-01 2.090 0.090927 .
## ect6 -5.568e-02 2.741e-02 -2.031 0.097964 .
## constant 5.916e+00 2.251e+01 0.263 0.803199
## MRO.dl1 -2.017e+00 2.753e-01 -7.330 0.000741 ***
## OMO.dl1 6.290e-04 3.083e-04 2.040 0.096886 .
## GDP.dl1 -1.782e+01 4.668e+00 -3.818 0.012395 *
## unemployment.dl1 2.468e-02 3.425e-01 0.072 0.945356
## loanshsannualgrowth.dl1 7.634e-02 7.211e-01 0.106 0.919807
## CCI.dl1 -8.749e-03 1.383e-02 -0.633 0.554678
## finalconsumption.dl1 -4.527e-06 8.402e-06 -0.539 0.613147
## MRO.dl2 -2.217e+00 4.277e-01 -5.183 0.003516 **
## OMO.dl2 1.896e-03 5.153e-04 3.680 0.014287 *
## GDP.dl2 -2.062e+01 6.181e+00 -3.336 0.020641 *
## unemployment.dl2 9.767e-01 6.507e-01 1.501 0.193644
## loanshsannualgrowth.dl2 4.963e-02 3.262e-01 0.152 0.885008
## CCI.dl2 9.857e-03 1.134e-02 0.869 0.424428
## finalconsumption.dl2 -4.141e-06 2.686e-06 -1.541 0.183846
## MRO.dl3 -6.739e-01 2.581e-01 -2.610 0.047644 *
## OMO.dl3 3.840e-03 7.319e-04 5.247 0.003335 **
## GDP.dl3 -1.746e+01 5.558e+00 -3.141 0.025644 *
## unemployment.dl3 8.378e-01 7.413e-01 1.130 0.309714
## loanshsannualgrowth.dl3 5.225e-01 2.501e-01 2.090 0.090950 .
## CCI.dl3 3.294e-03 2.016e-02 0.163 0.876644
## finalconsumption.dl3 1.156e-06 6.325e-06 0.183 0.862120
## MRO.dl4 5.528e-01 3.167e-01 1.745 0.141364
## OMO.dl4 3.661e-03 7.117e-04 5.144 0.003634 **
## GDP.dl4 -3.294e+01 8.844e+00 -3.725 0.013643 *
## unemployment.dl4 1.014e+00 8.352e-01 1.214 0.278861
## loanshsannualgrowth.dl4 4.453e-01 2.794e-01 1.594 0.171815
## CCI.dl4 -5.694e-03 1.256e-02 -0.453 0.669303
## finalconsumption.dl4 3.851e-06 8.696e-06 0.443 0.676409
## MRO.dl5 8.328e-01 2.553e-01 3.263 0.022377 *
## OMO.dl5 2.664e-03 7.959e-04 3.347 0.020392 *
## GDP.dl5 -4.087e+01 1.397e+01 -2.925 0.032807 *
## unemployment.dl5 1.773e-01 9.134e-01 0.194 0.853706
## loanshsannualgrowth.dl5 1.621e+00 6.265e-01 2.588 0.048945 *
## CCI.dl5 -2.285e-02 1.568e-02 -1.457 0.204853
## finalconsumption.dl5 6.298e-06 4.694e-06 1.342 0.237375
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.06643 on 5 degrees of freedom
## Multiple R-squared: 0.9907, Adjusted R-squared: 0.9127
## F-statistic: 12.7 on 42 and 5 DF, p-value: 0.004831
##
##
## Response OMO.d :
##
## Call:
## lm(formula = OMO.d ~ ect1 + ect2 + ect3 + ect4 + ect5 + ect6 +
## constant + MRO.dl1 + OMO.dl1 + GDP.dl1 + unemployment.dl1 +
## loanshsannualgrowth.dl1 + CCI.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + GDP.dl2 + unemployment.dl2 + loanshsannualgrowth.dl2 +
## CCI.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 + GDP.dl3 +
## unemployment.dl3 + loanshsannualgrowth.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + GDP.dl4 + unemployment.dl4 + loanshsannualgrowth.dl4 +
## CCI.dl4 + finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + GDP.dl5 +
## unemployment.dl5 + loanshsannualgrowth.dl5 + CCI.dl5 + finalconsumption.dl5 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7 8
## 3.0672 -10.3385 13.2260 -0.2493 -19.5512 7.2276 11.9175 -35.7842
## 9 10 11 12 13 14 15 16
## 21.2422 -4.3906 16.5952 -1.3381 34.5427 -23.7360 2.6872 -50.3794
## 17 18 19 20 21 22 23 24
## 43.6315 68.7425 -23.4961 1.0027 -56.8569 -11.6143 3.7606 23.6928
## 25 26 27 28 29 30 31 32
## -2.9595 -19.9590 15.6485 39.5119 -21.2734 -34.9730 1.9906 23.1773
## 33 34 35 36 37 38 39 40
## -6.8203 7.7085 -6.3829 -6.3629 19.3577 -12.7588 4.5367 3.0119
## 41 42 43 44 45 46 47
## -32.0653 35.7828 -27.7023 33.6120 -2.0952 20.9757 -45.5625
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 -1.299e+02 3.947e+02 -0.329 0.7554
## ect2 7.011e-01 1.374e+00 0.510 0.6315
## ect3 -2.363e+03 1.060e+04 -0.223 0.8324
## ect4 -1.490e+03 1.503e+03 -0.991 0.3672
## ect5 -7.447e+02 7.162e+02 -1.040 0.3461
## ect6 -3.075e+01 3.209e+01 -0.958 0.3819
## constant 2.995e+04 2.635e+04 1.136 0.3073
## MRO.dl1 1.338e+02 3.222e+02 0.415 0.6953
## OMO.dl1 -1.049e+00 3.610e-01 -2.905 0.0336 *
## GDP.dl1 6.268e+03 5.465e+03 1.147 0.3033
## unemployment.dl1 -5.336e+02 4.009e+02 -1.331 0.2407
## loanshsannualgrowth.dl1 -4.144e+02 8.442e+02 -0.491 0.6443
## CCI.dl1 1.484e+01 1.618e+01 0.917 0.4012
## finalconsumption.dl1 -5.460e-03 9.836e-03 -0.555 0.6027
## MRO.dl2 4.801e+02 5.007e+02 0.959 0.3817
## OMO.dl2 -7.596e-01 6.032e-01 -1.259 0.2635
## GDP.dl2 9.416e+03 7.236e+03 1.301 0.2499
## unemployment.dl2 -1.187e+03 7.617e+02 -1.558 0.1799
## loanshsannualgrowth.dl2 -3.296e+01 3.818e+02 -0.086 0.9346
## CCI.dl2 -4.880e+00 1.327e+01 -0.368 0.7282
## finalconsumption.dl2 2.911e-03 3.145e-03 0.926 0.3970
## MRO.dl3 -7.417e+01 3.022e+02 -0.245 0.8159
## OMO.dl3 -1.787e+00 8.568e-01 -2.086 0.0914 .
## GDP.dl3 4.945e+03 6.506e+03 0.760 0.4815
## unemployment.dl3 -1.205e+03 8.678e+02 -1.388 0.2237
## loanshsannualgrowth.dl3 -6.607e+02 2.927e+02 -2.257 0.0736 .
## CCI.dl3 4.159e+00 2.360e+01 0.176 0.8671
## finalconsumption.dl3 -3.866e-03 7.404e-03 -0.522 0.6239
## MRO.dl4 -7.917e+02 3.707e+02 -2.135 0.0858 .
## OMO.dl4 -1.252e+00 8.331e-01 -1.503 0.1931
## GDP.dl4 2.630e+03 1.035e+04 0.254 0.8096
## unemployment.dl4 -1.443e+03 9.777e+02 -1.476 0.2000
## loanshsannualgrowth.dl4 -5.533e+02 3.270e+02 -1.692 0.1515
## CCI.dl4 -2.449e+01 1.470e+01 -1.665 0.1567
## finalconsumption.dl4 -1.051e-02 1.018e-02 -1.032 0.3492
## MRO.dl5 -5.885e+02 2.988e+02 -1.969 0.1060
## OMO.dl5 -6.494e-01 9.317e-01 -0.697 0.5169
## GDP.dl5 1.121e+04 1.635e+04 0.685 0.5236
## unemployment.dl5 -1.266e+03 1.069e+03 -1.184 0.2895
## loanshsannualgrowth.dl5 -1.146e+03 7.334e+02 -1.563 0.1788
## CCI.dl5 -2.006e+01 1.835e+01 -1.093 0.3242
## finalconsumption.dl5 -8.993e-03 5.495e-03 -1.637 0.1626
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 77.76 on 5 degrees of freedom
## Multiple R-squared: 0.963, Adjusted R-squared: 0.6524
## F-statistic: 3.1 on 42 and 5 DF, p-value: 0.1034
##
##
## Response GDP.d :
##
## Call:
## lm(formula = GDP.d ~ ect1 + ect2 + ect3 + ect4 + ect5 + ect6 +
## constant + MRO.dl1 + OMO.dl1 + GDP.dl1 + unemployment.dl1 +
## loanshsannualgrowth.dl1 + CCI.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + GDP.dl2 + unemployment.dl2 + loanshsannualgrowth.dl2 +
## CCI.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 + GDP.dl3 +
## unemployment.dl3 + loanshsannualgrowth.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + GDP.dl4 + unemployment.dl4 + loanshsannualgrowth.dl4 +
## CCI.dl4 + finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + GDP.dl5 +
## unemployment.dl5 + loanshsannualgrowth.dl5 + CCI.dl5 + finalconsumption.dl5 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7
## -8.285e-05 7.196e-04 -1.295e-03 9.004e-04 -7.242e-05 -4.875e-05 2.230e-04
## 8 9 10 11 12 13 14
## 1.574e-04 1.477e-04 -2.154e-04 -1.064e-03 7.668e-04 -2.008e-03 -5.097e-05
## 15 16 17 18 19 20 21
## 3.852e-04 1.450e-03 2.429e-03 -7.242e-03 2.510e-03 -2.094e-03 3.739e-03
## 22 23 24 25 26 27 28
## 1.636e-03 -7.321e-04 -1.586e-03 6.695e-05 5.731e-04 1.161e-03 -2.379e-03
## 29 30 31 32 33 34 35
## 2.838e-05 2.106e-03 -3.499e-04 -3.475e-04 6.820e-04 -1.914e-03 4.554e-04
## 36 37 38 39 40 41 42
## 7.540e-04 9.807e-04 -1.901e-03 -1.504e-03 2.603e-03 5.719e-04 -1.469e-03
## 43 44 45 46 47
## -3.249e-04 6.504e-04 1.139e-03 -4.572e-03 4.416e-03
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 -6.230e-02 3.004e-02 -2.074 0.0928 .
## ect2 -1.964e-04 1.046e-04 -1.878 0.1191
## ect3 -3.816e-01 8.068e-01 -0.473 0.6562
## ect4 2.561e-01 1.144e-01 2.238 0.0754 .
## ect5 7.925e-02 5.451e-02 1.454 0.2057
## ect6 6.609e-03 2.442e-03 2.706 0.0425 *
## constant -3.605e+00 2.006e+00 -1.797 0.1322
## MRO.dl1 9.640e-03 2.453e-02 0.393 0.7105
## OMO.dl1 9.844e-06 2.747e-05 0.358 0.7348
## GDP.dl1 -2.920e-01 4.159e-01 -0.702 0.5140
## unemployment.dl1 8.902e-02 3.052e-02 2.917 0.0331 *
## loanshsannualgrowth.dl1 1.088e-01 6.425e-02 1.693 0.1512
## CCI.dl1 -7.739e-04 1.232e-03 -0.628 0.5574
## finalconsumption.dl1 1.871e-06 7.486e-07 2.500 0.0545 .
## MRO.dl2 1.165e-02 3.811e-02 0.306 0.7721
## OMO.dl2 -8.199e-05 4.591e-05 -1.786 0.1342
## GDP.dl2 -6.296e-01 5.507e-01 -1.143 0.3047
## unemployment.dl2 1.056e-01 5.797e-02 1.821 0.1283
## loanshsannualgrowth.dl2 -1.255e-02 2.906e-02 -0.432 0.6837
## CCI.dl2 -1.393e-03 1.010e-03 -1.379 0.2263
## finalconsumption.dl2 2.072e-07 2.393e-07 0.866 0.4262
## MRO.dl3 1.958e-02 2.300e-02 0.851 0.4335
## OMO.dl3 4.706e-05 6.521e-05 0.722 0.5028
## GDP.dl3 -1.614e+00 4.952e-01 -3.259 0.0225 *
## unemployment.dl3 1.405e-01 6.605e-02 2.126 0.0868 .
## loanshsannualgrowth.dl3 1.556e-02 2.228e-02 0.698 0.5162
## CCI.dl3 -3.125e-03 1.797e-03 -1.739 0.1425
## finalconsumption.dl3 1.236e-06 5.636e-07 2.193 0.0798 .
## MRO.dl4 1.236e-02 2.822e-02 0.438 0.6796
## OMO.dl4 -1.081e-04 6.341e-05 -1.705 0.1489
## GDP.dl4 -4.483e-01 7.880e-01 -0.569 0.5940
## unemployment.dl4 1.553e-01 7.441e-02 2.087 0.0912 .
## loanshsannualgrowth.dl4 6.029e-02 2.489e-02 2.422 0.0600 .
## CCI.dl4 8.652e-04 1.119e-03 0.773 0.4744
## finalconsumption.dl4 1.637e-06 7.749e-07 2.113 0.0883 .
## MRO.dl5 -6.564e-02 2.274e-02 -2.886 0.0343 *
## OMO.dl5 -1.075e-04 7.091e-05 -1.516 0.1899
## GDP.dl5 -1.602e+00 1.245e+00 -1.287 0.2543
## unemployment.dl5 2.232e-01 8.138e-02 2.742 0.0407 *
## loanshsannualgrowth.dl5 8.514e-02 5.582e-02 1.525 0.1877
## CCI.dl5 8.020e-04 1.397e-03 0.574 0.5908
## finalconsumption.dl5 5.314e-07 4.182e-07 1.271 0.2598
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.005919 on 5 degrees of freedom
## Multiple R-squared: 0.9824, Adjusted R-squared: 0.8348
## F-statistic: 6.655 on 42 and 5 DF, p-value: 0.02114
##
##
## Response unemployment.d :
##
## Call:
## lm(formula = unemployment.d ~ ect1 + ect2 + ect3 + ect4 + ect5 +
## ect6 + constant + MRO.dl1 + OMO.dl1 + GDP.dl1 + unemployment.dl1 +
## loanshsannualgrowth.dl1 + CCI.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + GDP.dl2 + unemployment.dl2 + loanshsannualgrowth.dl2 +
## CCI.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 + GDP.dl3 +
## unemployment.dl3 + loanshsannualgrowth.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + GDP.dl4 + unemployment.dl4 + loanshsannualgrowth.dl4 +
## CCI.dl4 + finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + GDP.dl5 +
## unemployment.dl5 + loanshsannualgrowth.dl5 + CCI.dl5 + finalconsumption.dl5 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.0062732 0.0097675 -0.0313398 0.0348057 0.0154670 -0.0345319 0.0240909
## 8 9 10 11 12 13 14
## 0.0258023 -0.0261837 -0.0388022 -0.0119950 -0.0107685 -0.0003642 0.0133640
## 15 16 17 18 19 20 21
## -0.0317092 0.0175248 -0.0447077 -0.0495551 0.0972761 -0.0088640 0.0496258
## 22 23 24 25 26 27 28
## 0.0023810 -0.0035000 -0.0450111 -0.0025866 0.0035860 -0.0286335 0.0267613
## 29 30 31 32 33 34 35
## 0.0406944 -0.0045963 -0.0338454 -0.0012846 0.0656169 -0.0753368 0.0124959
## 36 37 38 39 40 41 42
## -0.0021818 0.0503039 -0.0035263 -0.1085942 0.0739356 0.0442580 -0.1198794
## 43 44 45 46 47
## 0.0322643 -0.0299483 0.0902944 -0.0506113 0.0617677
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 5.252e-01 6.981e-01 0.752 0.4858
## ect2 1.081e-03 2.430e-03 0.445 0.6749
## ect3 -2.569e+01 1.875e+01 -1.370 0.2290
## ect4 1.334e-01 2.658e+00 0.050 0.9619
## ect5 9.454e-01 1.267e+00 0.746 0.4890
## ect6 -5.177e-02 5.675e-02 -0.912 0.4035
## constant 3.101e+01 4.661e+01 0.665 0.5353
## MRO.dl1 -2.503e-01 5.699e-01 -0.439 0.6789
## OMO.dl1 -1.550e-04 6.384e-04 -0.243 0.8178
## GDP.dl1 -5.204e+00 9.665e+00 -0.538 0.6134
## unemployment.dl1 -6.038e-01 7.091e-01 -0.852 0.4333
## loanshsannualgrowth.dl1 1.162e+00 1.493e+00 0.779 0.4714
## CCI.dl1 -1.268e-02 2.862e-02 -0.443 0.6763
## finalconsumption.dl1 9.171e-06 1.740e-05 0.527 0.6206
## MRO.dl2 -8.802e-01 8.855e-01 -0.994 0.3659
## OMO.dl2 -9.722e-04 1.067e-03 -0.911 0.4039
## GDP.dl2 -1.344e+01 1.280e+01 -1.050 0.3417
## unemployment.dl2 -3.672e-01 1.347e+00 -0.273 0.7961
## loanshsannualgrowth.dl2 2.941e-01 6.753e-01 0.436 0.6813
## CCI.dl2 -2.783e-02 2.348e-02 -1.185 0.2891
## finalconsumption.dl2 -2.363e-06 5.562e-06 -0.425 0.6885
## MRO.dl3 -6.372e-01 5.345e-01 -1.192 0.2867
## OMO.dl3 1.359e-03 1.515e-03 0.897 0.4108
## GDP.dl3 -8.657e+00 1.151e+01 -0.752 0.4858
## unemployment.dl3 -1.227e-01 1.535e+00 -0.080 0.9394
## loanshsannualgrowth.dl3 -5.757e-01 5.177e-01 -1.112 0.3167
## CCI.dl3 -5.603e-02 4.175e-02 -1.342 0.2373
## finalconsumption.dl3 -9.035e-07 1.310e-05 -0.069 0.9477
## MRO.dl4 5.008e-01 6.557e-01 0.764 0.4795
## OMO.dl4 1.903e-03 1.473e-03 1.291 0.2531
## GDP.dl4 -1.847e+01 1.831e+01 -1.009 0.3593
## unemployment.dl4 -3.648e-01 1.729e+00 -0.211 0.8412
## loanshsannualgrowth.dl4 -1.010e-01 5.784e-01 -0.175 0.8683
## CCI.dl4 -6.176e-02 2.600e-02 -2.375 0.0636 .
## finalconsumption.dl4 4.432e-06 1.801e-05 0.246 0.8153
## MRO.dl5 5.509e-01 5.285e-01 1.042 0.3450
## OMO.dl5 1.938e-03 1.648e-03 1.176 0.2925
## GDP.dl5 -3.709e+01 2.892e+01 -1.283 0.2559
## unemployment.dl5 -4.175e-01 1.891e+00 -0.221 0.8340
## loanshsannualgrowth.dl5 6.424e-01 1.297e+00 0.495 0.6414
## CCI.dl5 -7.502e-02 3.246e-02 -2.311 0.0688 .
## finalconsumption.dl5 -1.588e-06 9.718e-06 -0.163 0.8766
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1375 on 5 degrees of freedom
## Multiple R-squared: 0.956, Adjusted R-squared: 0.5865
## F-statistic: 2.587 on 42 and 5 DF, p-value: 0.1448
##
##
## Response loanshsannualgrowth.d :
##
## Call:
## lm(formula = loanshsannualgrowth.d ~ ect1 + ect2 + ect3 + ect4 +
## ect5 + ect6 + constant + MRO.dl1 + OMO.dl1 + GDP.dl1 + unemployment.dl1 +
## loanshsannualgrowth.dl1 + CCI.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + GDP.dl2 + unemployment.dl2 + loanshsannualgrowth.dl2 +
## CCI.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 + GDP.dl3 +
## unemployment.dl3 + loanshsannualgrowth.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + GDP.dl4 + unemployment.dl4 + loanshsannualgrowth.dl4 +
## CCI.dl4 + finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + GDP.dl5 +
## unemployment.dl5 + loanshsannualgrowth.dl5 + CCI.dl5 + finalconsumption.dl5 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.0027823 0.0018151 0.0023853 -0.0054298 -0.0017366 0.0104097 -0.0075512
## 8 9 10 11 12 13 14
## -0.0060168 0.0076353 0.0095406 -0.0053531 0.0006276 -0.0060664 -0.0038817
## 15 16 17 18 19 20 21
## 0.0197403 -0.0031300 0.0118633 -0.0339138 0.0007797 0.0074866 0.0028411
## 22 23 24 25 26 27 28
## 0.0062102 -0.0073681 0.0040221 0.0049834 -0.0006805 0.0039055 -0.0161488
## 29 30 31 32 33 34 35
## -0.0028249 0.0135913 0.0043436 -0.0054294 -0.0140611 0.0137635 0.0039955
## 36 37 38 39 40 41 42
## -0.0049765 -0.0071784 -0.0035740 0.0234063 -0.0103099 -0.0159641 0.0340281
## 43 44 45 46 47
## -0.0155394 0.0031423 -0.0131650 -0.0073398 0.0099051
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 -3.582e-01 1.762e-01 -2.033 0.0977 .
## ect2 7.132e-04 6.132e-04 1.163 0.2973
## ect3 -9.849e-01 4.731e+00 -0.208 0.8433
## ect4 3.037e-02 6.708e-01 0.045 0.9656
## ect5 -4.916e-01 3.197e-01 -1.538 0.1847
## ect6 5.776e-03 1.432e-02 0.403 0.7034
## constant -1.581e+00 1.176e+01 -0.134 0.8983
## MRO.dl1 -4.892e-02 1.438e-01 -0.340 0.7476
## OMO.dl1 3.965e-04 1.611e-04 2.461 0.0571 .
## GDP.dl1 1.931e+00 2.439e+00 0.792 0.4644
## unemployment.dl1 -7.904e-02 1.790e-01 -0.442 0.6772
## loanshsannualgrowth.dl1 -5.214e-01 3.768e-01 -1.384 0.2250
## CCI.dl1 4.846e-03 7.224e-03 0.671 0.5321
## finalconsumption.dl1 2.459e-06 4.390e-06 0.560 0.5995
## MRO.dl2 -2.922e-01 2.235e-01 -1.307 0.2480
## OMO.dl2 -2.725e-04 2.692e-04 -1.012 0.3580
## GDP.dl2 -2.044e+00 3.230e+00 -0.633 0.5547
## unemployment.dl2 -5.659e-02 3.400e-01 -0.166 0.8743
## loanshsannualgrowth.dl2 -2.952e-01 1.704e-01 -1.732 0.1438
## CCI.dl2 9.290e-03 5.925e-03 1.568 0.1777
## finalconsumption.dl2 -4.436e-06 1.404e-06 -3.161 0.0251 *
## MRO.dl3 -4.566e-01 1.349e-01 -3.386 0.0196 *
## OMO.dl3 5.230e-04 3.824e-04 1.368 0.2297
## GDP.dl3 -1.271e+00 2.904e+00 -0.438 0.6800
## unemployment.dl3 1.025e-01 3.873e-01 0.265 0.8019
## loanshsannualgrowth.dl3 -3.893e-01 1.307e-01 -2.980 0.0308 *
## CCI.dl3 2.867e-03 1.054e-02 0.272 0.7964
## finalconsumption.dl3 1.531e-06 3.305e-06 0.463 0.6626
## MRO.dl4 -2.355e-01 1.655e-01 -1.423 0.2140
## OMO.dl4 4.311e-04 3.719e-04 1.159 0.2987
## GDP.dl4 4.024e+00 4.621e+00 0.871 0.4237
## unemployment.dl4 1.374e-01 4.364e-01 0.315 0.7656
## loanshsannualgrowth.dl4 -5.807e-01 1.460e-01 -3.978 0.0105 *
## CCI.dl4 1.543e-02 6.563e-03 2.351 0.0655 .
## finalconsumption.dl4 1.593e-06 4.544e-06 0.351 0.7402
## MRO.dl5 -3.715e-01 1.334e-01 -2.785 0.0387 *
## OMO.dl5 8.015e-04 4.158e-04 1.927 0.1119
## GDP.dl5 -5.020e+00 7.299e+00 -0.688 0.5222
## unemployment.dl5 -8.820e-03 4.773e-01 -0.018 0.9860
## loanshsannualgrowth.dl5 -5.283e-01 3.274e-01 -1.614 0.1675
## CCI.dl5 -1.889e-03 8.192e-03 -0.231 0.8268
## finalconsumption.dl5 -2.725e-06 2.453e-06 -1.111 0.3171
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.03471 on 5 degrees of freedom
## Multiple R-squared: 0.9967, Adjusted R-squared: 0.9685
## F-statistic: 35.46 on 42 and 5 DF, p-value: 0.0004095
##
##
## Response CCI.d :
##
## Call:
## lm(formula = CCI.d ~ ect1 + ect2 + ect3 + ect4 + ect5 + ect6 +
## constant + MRO.dl1 + OMO.dl1 + GDP.dl1 + unemployment.dl1 +
## loanshsannualgrowth.dl1 + CCI.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + GDP.dl2 + unemployment.dl2 + loanshsannualgrowth.dl2 +
## CCI.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 + GDP.dl3 +
## unemployment.dl3 + loanshsannualgrowth.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + GDP.dl4 + unemployment.dl4 + loanshsannualgrowth.dl4 +
## CCI.dl4 + finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + GDP.dl5 +
## unemployment.dl5 + loanshsannualgrowth.dl5 + CCI.dl5 + finalconsumption.dl5 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7 8
## 0.02089 -0.21171 0.49789 -0.25157 -0.12497 0.37558 -0.20852 -0.63640
## 9 10 11 12 13 14 15 16
## 0.38353 0.04043 0.08274 -0.53369 0.92117 -0.32737 0.89337 -1.37974
## 17 18 19 20 21 22 23 24
## -0.64623 1.15324 0.51403 1.41790 -1.56577 -0.58509 -0.35640 0.73908
## 25 26 27 28 29 30 31 32
## 0.22496 -0.65867 -0.54259 1.13948 0.31976 -0.65214 -0.18291 0.18384
## 33 34 35 36 37 38 39 40
## -0.44029 0.89286 0.08390 -0.86900 0.28010 0.22687 0.80018 -0.89526
## 41 42 43 44 45 46 47
## -1.25824 1.80795 -1.09068 0.13665 -0.12011 1.18133 -0.78035
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 3.273e+01 1.171e+01 2.794 0.0383 *
## ect2 1.039e-01 4.077e-02 2.550 0.0513 .
## ect3 -7.810e+02 3.146e+02 -2.483 0.0557 .
## ect4 -6.799e+01 4.460e+01 -1.524 0.1879
## ect5 3.571e+00 2.125e+01 0.168 0.8732
## ect6 -2.861e+00 9.523e-01 -3.004 0.0300 *
## constant 2.079e+03 7.821e+02 2.658 0.0450 *
## MRO.dl1 -8.725e+00 9.562e+00 -0.912 0.4034
## OMO.dl1 -4.854e-03 1.071e-02 -0.453 0.6694
## GDP.dl1 -2.953e+01 1.622e+02 -0.182 0.8627
## unemployment.dl1 -2.176e+01 1.190e+01 -1.829 0.1270
## loanshsannualgrowth.dl1 -2.690e+01 2.505e+01 -1.074 0.3319
## CCI.dl1 -1.995e-01 4.803e-01 -0.415 0.6950
## finalconsumption.dl1 -5.763e-04 2.919e-04 -1.974 0.1053
## MRO.dl2 -7.523e+00 1.486e+01 -0.506 0.6342
## OMO.dl2 3.780e-02 1.790e-02 2.112 0.0884 .
## GDP.dl2 -1.849e+02 2.147e+02 -0.861 0.4285
## unemployment.dl2 -4.146e+01 2.260e+01 -1.834 0.1261
## loanshsannualgrowth.dl2 1.545e+01 1.133e+01 1.363 0.2310
## CCI.dl2 -8.407e-01 3.939e-01 -2.134 0.0859 .
## finalconsumption.dl2 -1.401e-04 9.331e-05 -1.501 0.1937
## MRO.dl3 -1.158e+01 8.968e+00 -1.292 0.2529
## OMO.dl3 1.303e-02 2.543e-02 0.512 0.6302
## GDP.dl3 -1.014e+02 1.931e+02 -0.525 0.6218
## unemployment.dl3 -4.158e+01 2.575e+01 -1.615 0.1673
## loanshsannualgrowth.dl3 3.828e+00 8.687e+00 0.441 0.6778
## CCI.dl3 3.096e-01 7.005e-01 0.442 0.6770
## finalconsumption.dl3 -3.167e-04 2.197e-04 -1.441 0.2091
## MRO.dl4 -8.750e+00 1.100e+01 -0.795 0.4625
## OMO.dl4 5.290e-02 2.472e-02 2.140 0.0854 .
## GDP.dl4 -5.899e+02 3.072e+02 -1.920 0.1129
## unemployment.dl4 -4.373e+01 2.901e+01 -1.507 0.1921
## loanshsannualgrowth.dl4 -2.874e-01 9.705e+00 -0.030 0.9775
## CCI.dl4 -1.053e+00 4.363e-01 -2.414 0.0606 .
## finalconsumption.dl4 -6.414e-04 3.021e-04 -2.123 0.0872 .
## MRO.dl5 1.289e+01 8.867e+00 1.454 0.2058
## OMO.dl5 6.897e-02 2.765e-02 2.495 0.0548 .
## GDP.dl5 -1.924e+02 4.853e+02 -0.396 0.7081
## unemployment.dl5 -5.642e+01 3.173e+01 -1.778 0.1355
## loanshsannualgrowth.dl5 -1.377e+01 2.176e+01 -0.633 0.5548
## CCI.dl5 -1.082e+00 5.447e-01 -1.987 0.1037
## finalconsumption.dl5 -2.549e-04 1.631e-04 -1.564 0.1787
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.308 on 5 degrees of freedom
## Multiple R-squared: 0.9258, Adjusted R-squared: 0.3026
## F-statistic: 1.486 on 42 and 5 DF, p-value: 0.354
##
##
## Response finalconsumption.d :
##
## Call:
## lm(formula = finalconsumption.d ~ ect1 + ect2 + ect3 + ect4 +
## ect5 + ect6 + constant + MRO.dl1 + OMO.dl1 + GDP.dl1 + unemployment.dl1 +
## loanshsannualgrowth.dl1 + CCI.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + GDP.dl2 + unemployment.dl2 + loanshsannualgrowth.dl2 +
## CCI.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 + GDP.dl3 +
## unemployment.dl3 + loanshsannualgrowth.dl3 + CCI.dl3 + finalconsumption.dl3 +
## MRO.dl4 + OMO.dl4 + GDP.dl4 + unemployment.dl4 + loanshsannualgrowth.dl4 +
## CCI.dl4 + finalconsumption.dl4 + MRO.dl5 + OMO.dl5 + GDP.dl5 +
## unemployment.dl5 + loanshsannualgrowth.dl5 + CCI.dl5 + finalconsumption.dl5 -
## 1, data = data.mat)
##
## Residuals:
## 1 2 3 4 5 6 7 8
## -200.28 -245.85 932.15 -982.80 -222.27 1008.12 -774.82 -734.01
## 9 10 11 12 13 14 15 16
## 696.90 1008.47 147.62 -229.22 474.98 -389.01 1505.42 -1057.97
## 17 18 19 20 21 22 23 24
## 168.08 955.30 -1568.82 1517.22 -1773.61 -275.92 -222.39 1276.95
## 25 26 27 28 29 30 31 32
## 355.01 -362.31 -22.01 -251.25 -477.85 71.56 656.92 -202.94
## 33 34 35 36 37 38 39 40
## -1721.03 2283.31 23.07 -703.51 -1185.17 533.85 3005.55 -2291.57
## 41 42 43 44 45 46 47
## -1757.19 3917.36 -1272.56 193.56 -1970.82 1726.49 -1562.74
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 5.452e+04 1.968e+04 2.771 0.0393 *
## ect2 1.006e+02 6.849e+01 1.469 0.2019
## ect3 2.040e+05 5.285e+05 0.386 0.7154
## ect4 -1.810e+05 7.493e+04 -2.416 0.0604 .
## ect5 -3.262e+04 3.571e+04 -0.913 0.4029
## ect6 -2.824e+03 1.600e+03 -1.765 0.1378
## constant 2.769e+06 1.314e+06 2.107 0.0889 .
## MRO.dl1 1.548e+04 1.607e+04 0.964 0.3795
## OMO.dl1 1.761e+01 1.800e+01 0.978 0.3728
## GDP.dl1 4.143e+05 2.725e+05 1.521 0.1888
## unemployment.dl1 -4.596e+04 1.999e+04 -2.299 0.0699 .
## loanshsannualgrowth.dl1 -2.725e+04 4.209e+04 -0.648 0.5459
## CCI.dl1 1.123e+03 8.069e+02 1.392 0.2228
## finalconsumption.dl1 -9.918e-01 4.904e-01 -2.022 0.0991 .
## MRO.dl2 3.708e+04 2.496e+04 1.485 0.1976
## OMO.dl2 3.757e+01 3.007e+01 1.249 0.2669
## GDP.dl2 5.102e+05 3.608e+05 1.414 0.2164
## unemployment.dl2 -8.940e+04 3.798e+04 -2.354 0.0652 .
## loanshsannualgrowth.dl2 8.324e+03 1.904e+04 0.437 0.6801
## CCI.dl2 1.586e+03 6.618e+02 2.396 0.0619 .
## finalconsumption.dl2 -4.164e-01 1.568e-01 -2.656 0.0451 *
## MRO.dl3 1.871e+04 1.507e+04 1.242 0.2693
## OMO.dl3 -4.144e+01 4.272e+01 -0.970 0.3766
## GDP.dl3 2.081e+05 3.244e+05 0.642 0.5494
## unemployment.dl3 -1.086e+05 4.327e+04 -2.509 0.0539 .
## loanshsannualgrowth.dl3 3.484e+04 1.459e+04 2.387 0.0626 .
## CCI.dl3 2.006e+03 1.177e+03 1.704 0.1491
## finalconsumption.dl3 -1.135e+00 3.692e-01 -3.074 0.0277 *
## MRO.dl4 -5.400e+03 1.848e+04 -0.292 0.7819
## OMO.dl4 -2.033e+01 4.154e+01 -0.490 0.6452
## GDP.dl4 7.391e+05 5.162e+05 1.432 0.2116
## unemployment.dl4 -1.191e+05 4.874e+04 -2.443 0.0584 .
## loanshsannualgrowth.dl4 2.086e+03 1.631e+04 0.128 0.9032
## CCI.dl4 5.680e+02 7.331e+02 0.775 0.4735
## finalconsumption.dl4 -1.126e+00 5.076e-01 -2.218 0.0774 .
## MRO.dl5 2.534e+04 1.490e+04 1.701 0.1497
## OMO.dl5 4.166e+01 4.645e+01 0.897 0.4109
## GDP.dl5 1.109e+06 8.153e+05 1.361 0.2317
## unemployment.dl5 -1.453e+05 5.331e+04 -2.726 0.0415 *
## loanshsannualgrowth.dl5 -3.113e+04 3.657e+04 -0.851 0.4335
## CCI.dl5 -7.093e+02 9.151e+02 -0.775 0.4733
## finalconsumption.dl5 -7.294e-01 2.740e-01 -2.662 0.0448 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3877 on 5 degrees of freedom
## Multiple R-squared: 0.9962, Adjusted R-squared: 0.9643
## F-statistic: 31.2 on 42 and 5 DF, p-value: 0.0005595## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 6 6 1 6
##
## $criteria
## 1 2 3 4 5
## AIC(n) 2.082059e+01 2.089102e+01 2.077284e+01 2.047392e+01 2.044448e+01
## HQ(n) 2.090994e+01 2.103994e+01 2.098132e+01 2.074197e+01 2.077209e+01
## SC(n) 2.105911e+01 2.128856e+01 2.132938e+01 2.118948e+01 2.131904e+01
## FPE(n) 1.102626e+09 1.184692e+09 1.055840e+09 7.873183e+08 7.711244e+08
## 6 7 8 9 10
## AIC(n) 2.022831e+01 2.022932e+01 2.039105e+01 2.053559e+01 2.043323e+01
## HQ(n) 2.061549e+01 2.067607e+01 2.089737e+01 2.110148e+01 2.105869e+01
## SC(n) 2.126189e+01 2.142191e+01 2.174265e+01 2.204621e+01 2.210286e+01
## FPE(n) 6.291822e+08 6.411304e+08 7.719766e+08 9.205840e+08 8.654678e+08##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 0.35979669 0.07847159
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | 4.09 6.50 8.18 11.65
## r = 0 | 26.38 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## MRO.l6 totalassetshs_diff.l6
## MRO.l6 1.00000e+00 1.000000e+00
## totalassetshs_diff.l6 3.71072e-05 -6.569614e-06
##
## Weights W:
## (This is the loading matrix)
##
## MRO.l6 totalassetshs_diff.l6
## MRO.d -1.993742e-02 -4.691205e-03
## totalassetshs_diff.d -1.588643e+03 1.859275e+04
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 10 1 1 1
##
## $criteria
## 1 2 3 4 5
## AIC(n) 3.359993e+01 3.371064e+01 3.383363e+01 3.390122e+01 3.392050e+01
## HQ(n) 3.369092e+01 3.386229e+01 3.404594e+01 3.417419e+01 3.425413e+01
## SC(n) 3.384817e+01 3.412437e+01 3.441285e+01 3.464593e+01 3.483071e+01
## FPE(n) 3.912695e+14 4.378539e+14 4.971470e+14 5.357622e+14 5.525095e+14
## 6 7 8 9 10
## AIC(n) 3.396983e+01 3.397575e+01 3.405351e+01 3.377214e+01 3.344911e+01
## HQ(n) 3.436412e+01 3.443070e+01 3.456911e+01 3.434841e+01 3.408603e+01
## SC(n) 3.504553e+01 3.521694e+01 3.546019e+01 3.534432e+01 3.518678e+01
## FPE(n) 5.903944e+14 6.082506e+14 6.790952e+14 5.350473e+14 4.096535e+14##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 0.191592303 0.001016666
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | 0.05 6.50 8.18 11.65
## r = 0 | 9.83 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## OMO.l6 totalassetshs_diff.l6
## OMO.l6 1.000000000 1.000000e+00
## totalassetshs_diff.l6 0.009085887 6.022359e-05
##
## Weights W:
## (This is the loading matrix)
##
## OMO.l6 totalassetshs_diff.l6
## OMO.d 0.01532957 -0.005753946
## totalassetshs_diff.d -47.37345656 -2.113000291
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 2 2 2 2
##
## $criteria
## 1 2 3 4 5
## AIC(n) 2.199085e+01 2.132519e+01 2.140307e+01 2.149492e+01 2.157531e+01
## HQ(n) 2.207973e+01 2.147332e+01 2.161046e+01 2.176156e+01 2.190121e+01
## SC(n) 2.222704e+01 2.171884e+01 2.195418e+01 2.220349e+01 2.244134e+01
## FPE(n) 3.553505e+09 1.828575e+09 1.982329e+09 2.184154e+09 2.386256e+09
## 6 7 8 9 10
## AIC(n) 2.159642e+01 2.151631e+01 2.153409e+01 2.164587e+01 2.158524e+01
## HQ(n) 2.198156e+01 2.196071e+01 2.203774e+01 2.220878e+01 2.220740e+01
## SC(n) 2.261990e+01 2.269725e+01 2.287250e+01 2.314174e+01 2.323857e+01
## FPE(n) 2.466340e+09 2.314618e+09 2.409424e+09 2.774549e+09 2.711676e+09##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 0.29536481 0.05366998
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | 3.03 6.50 8.18 11.65
## r = 0 | 22.29 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## hsassets.l2 Finalconsumptionmag.l2
## hsassets.l2 1 1
## Finalconsumptionmag.l2 -18227533 -244817
##
## Weights W:
## (This is the loading matrix)
##
## hsassets.l2 Finalconsumptionmag.l2
## hsassets.d -7.466474e-05 -2.446995e-02
## Finalconsumptionmag.d 2.888021e-08 9.395244e-08
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 8 7 7 10
##
## $criteria
## 1 2 3 4 5
## AIC(n) 2.142461e+01 2.124899e+01 2.147031e+01 2.119352e+01 2.121868e+01
## HQ(n) 2.148938e+01 2.135693e+01 2.162143e+01 2.138783e+01 2.145617e+01
## SC(n) 2.172304e+01 2.174638e+01 2.216666e+01 2.208883e+01 2.231295e+01
## FPE(n) 2.024409e+09 1.723499e+09 2.225545e+09 1.801381e+09 2.067795e+09
## 6 7 8 9 10
## AIC(n) 2.112519e+01 1.779746e+01 1.776988e+01 NaN NaN
## HQ(n) 2.140585e+01 1.812130e+01 1.813690e+01 NaN NaN
## SC(n) 2.241840e+01 1.928964e+01 1.946101e+01 NaN NaN
## FPE(n) 2.269518e+09 1.108669e+08 1.847235e+08 -6.261283e-08 -Inf##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 1.000000000 0.007983767
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | 0.18 6.50 8.18 11.65
## r = 0 | NaN 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## hsassets.l9 durablehs.l9
## hsassets.l9 1.0 1.0
## durablehs.l9 803485.4 -267127.7
##
## Weights W:
## (This is the loading matrix)
##
## hsassets.l9 durablehs.l9
## hsassets.d -9.761049e-02 -4.924270e-02
## durablehs.d -7.599131e-07 -6.529027e-08
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 3 2 2 3
##
## $criteria
## 1 2 3 4 5
## AIC(n) 2.643023e+01 2.558036e+01 2.556901e+01 2.558227e+01 2.571113e+01
## HQ(n) 2.651911e+01 2.572849e+01 2.577639e+01 2.584891e+01 2.603702e+01
## SC(n) 2.666642e+01 2.597401e+01 2.612012e+01 2.629083e+01 2.657715e+01
## FPE(n) 3.010596e+11 1.288581e+11 1.277668e+11 1.301354e+11 1.492373e+11
## 6 7 8 9 10
## AIC(n) 2.577692e+01 2.564743e+01 2.577236e+01 2.580693e+01 2.588136e+01
## HQ(n) 2.616206e+01 2.609183e+01 2.627601e+01 2.636984e+01 2.650351e+01
## SC(n) 2.680041e+01 2.682838e+01 2.711076e+01 2.730280e+01 2.753468e+01
## FPE(n) 1.612957e+11 1.440796e+11 1.669433e+11 1.779581e+11 1.990747e+11##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 0.1776930 0.0425294
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | 2.39 6.50 8.18 11.65
## r = 0 | 13.15 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## hsassets.l2 CCI.l2
## hsassets.l2 1 1.00
## CCI.l2 -1033094 17085.32
##
## Weights W:
## (This is the loading matrix)
##
## hsassets.l2 CCI.l2
## hsassets.d 4.665975e-04 -2.028839e-02
## CCI.d 1.806019e-07 8.396235e-08
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 7 1 1 7
##
## $criteria
## 1 2 3 4 5 6
## AIC(n) -0.26054546 -0.23078217 -0.11299383 -0.08187188 -0.01304564 -0.1835513
## HQ(n) -0.17166593 -0.08264963 0.09439174 0.18476671 0.31284596 0.2015933
## SC(n) -0.02435641 0.16286625 0.43811397 0.62669529 0.85298089 0.8399346
## FPE(n) 0.77089903 0.79519661 0.89715259 0.93025339 1.00464303 0.8573010
## 7 8 9 10
## AIC(n) -0.52998485 -0.43484747 -0.3434521 -0.4099794
## HQ(n) -0.08558721 0.06880319 0.2194516 0.2121773
## SC(n) 0.65096043 0.90355718 1.1524119 1.2433440
## FPE(n) 0.61644657 0.69330547 0.7822562 0.7600348##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 0.2905202 0.1044079
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | 6.06 6.50 8.18 11.65
## r = 0 | 24.94 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## CCI.l2 Finalconsumptionmag.l2
## CCI.l2 1.0000 1.0000000
## Finalconsumptionmag.l2 141.2335 -0.7701332
##
## Weights W:
## (This is the loading matrix)
##
## CCI.l2 Finalconsumptionmag.l2
## CCI.d -0.002259561 -0.168545342
## Finalconsumptionmag.d -0.003716164 0.009575811
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 10 10 10 9
##
## $criteria
## 1 2 3 4 5 6
## AIC(n) -2.37825252 -2.59870083 -2.35593605 -2.58657481 -2.85645211 -2.92909776
## HQ(n) -2.31348449 -2.49075411 -2.20481065 -2.39227072 -2.61896933 -2.64843629
## SC(n) -2.07981754 -2.10130919 -1.65958776 -1.69126987 -1.76219050 -1.63587950
## FPE(n) 0.09307804 0.07576899 0.09996101 0.08473055 0.07241161 0.08115046
## 7 8 9 10
## AIC(n) -4.00590955 -5.7554281 -9.310894323 -Inf
## HQ(n) -3.68206939 -5.3884093 -8.900696795 -Inf
## SC(n) -2.51373464 -4.0642965 -7.420806103 -Inf
## FPE(n) 0.03764582 0.0112101 0.000969745 NaN##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 1 1
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | NaN 6.50 8.18 11.65
## r = 0 | NaN 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## CCI.l10 durablehs.l10
## CCI.l10 1.000000 1.000000
## durablehs.l10 -1.844122 4.576057
##
## Weights W:
## (This is the loading matrix)
##
## CCI.l10 durablehs.l10
## CCI.d -0.82063061 0.12805410
## durablehs.d -0.03821304 -0.06964912
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 9 3 2 9
##
## $criteria
## 1 2 3 4 5 6
## AIC(n) -1.4177349 -1.9409118 -2.0374933 -2.0255767 -1.967462 -2.0911907
## HQ(n) -1.3303618 -1.7952900 -1.8336227 -1.7634573 -1.647094 -1.7125739
## SC(n) -1.1882921 -1.5585072 -1.5021269 -1.3372484 -1.126172 -1.0969387
## FPE(n) 0.2423321 0.1437657 0.1308386 0.1329681 0.141871 0.1265957
## 7 8 9 10
## AIC(n) -2.0847115 -2.1103864 -2.2209903 -2.1757848
## HQ(n) -1.6478459 -1.6152721 -1.6676272 -1.5641730
## SC(n) -0.9374977 -0.8102107 -0.7678528 -0.5696855
## FPE(n) 0.1291691 0.1282189 0.1175664 0.1268162##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 0.13575442 0.04684662
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | 2.73 6.50 8.18 11.65
## r = 0 | 11.05 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## loanshsannualgrowth.l3 CCI.l3
## loanshsannualgrowth.l3 1.0000000 1.0000000
## CCI.l3 -0.7558389 0.1354641
##
## Weights W:
## (This is the loading matrix)
##
## loanshsannualgrowth.l3 CCI.l3
## loanshsannualgrowth.d -0.008478487 -0.0100707
## CCI.d 0.112638035 -0.3765174
##
## Optimal (m+1)-segment partition:
##
## Call:
## breakpoints.formula(formula = hsassets ~ 1, data = dataeurozone)
##
## Breakpoints at observation number:
##
## m = 1 12
## m = 2 11 46
## m = 3 11 38 47
## m = 4 11 28 37 47
## m = 5 9 17 28 37 47
## m = 6 9 17 25 33 41 49
##
## Corresponding to breakdates:
##
## m = 1 0.210526315789474
## m = 2 0.192982456140351
## m = 3 0.192982456140351
## m = 4 0.192982456140351 0.491228070175439 0.649122807017544
## m = 5 0.157894736842105 0.298245614035088 0.491228070175439 0.649122807017544
## m = 6 0.157894736842105 0.298245614035088 0.43859649122807 0.578947368421053
##
## m = 1
## m = 2 0.807017543859649
## m = 3 0.666666666666667 0.824561403508772
## m = 4 0.824561403508772
## m = 5 0.824561403508772
## m = 6 0.719298245614035 0.859649122807018
##
## Fit:
##
## m 0 1 2 3 4 5 6
## RSS 1.299e+14 5.657e+13 1.929e+13 1.583e+13 1.273e+13 8.358e+12 1.175e+13
## BIC 1.792e+03 1.752e+03 1.699e+03 1.696e+03 1.692e+03 1.676e+03 1.703e+03##
## Optimal (m+1)-segment partition:
##
## Call:
## breakpoints.formula(formula = CCI ~ 1, data = dataeurozone)
##
## Breakpoints at observation number:
##
## m = 1 26
## m = 2 26 37
## m = 3 14 26 37
## m = 4 14 26 37 50
## m = 5 9 18 27 37 50
##
## Corresponding to breakdates:
##
## m = 1 0.433333333333333
## m = 2 0.433333333333333 0.616666666666667
## m = 3 0.233333333333333 0.433333333333333 0.616666666666667
## m = 4 0.233333333333333 0.433333333333333 0.616666666666667
## m = 5 0.15 0.3 0.45 0.616666666666667
##
## m = 1
## m = 2
## m = 3
## m = 4 0.833333333333333
## m = 5 0.833333333333333
##
## Fit:
##
## m 0 1 2 3 4 5
## RSS 2303.8 1461.2 918.7 575.2 519.5 870.7
## BIC 397.3 378.2 358.6 338.6 340.7 379.9
##
## Optimal (m+1)-segment partition:
##
## Call:
## breakpoints.formula(formula = interactiontermassetCCI ~ 1)
##
## Breakpoints at observation number:
##
## m = 1 26
## m = 2 26 37
## m = 3 14 26 37
## m = 4 14 26 37 49
## m = 5 8 16 26 37 49
## m = 6 8 16 25 33 41 49
##
## Corresponding to breakdates:
##
## m = 1 0.456140350877193
## m = 2 0.456140350877193 0.649122807017544
## m = 3 0.245614035087719 0.456140350877193 0.649122807017544
## m = 4 0.245614035087719 0.456140350877193 0.649122807017544
## m = 5 0.140350877192982 0.280701754385965 0.456140350877193 0.649122807017544
## m = 6 0.140350877192982 0.280701754385965 0.43859649122807 0.578947368421053
##
## m = 1
## m = 2
## m = 3
## m = 4 0.859649122807018
## m = 5 0.859649122807018
## m = 6 0.719298245614035 0.859649122807018
##
## Fit:
##
## m 0 1 2 3 4 5 6
## RSS 2.000e+18 1.034e+18 6.862e+17 4.535e+17 4.216e+17 4.226e+17 6.133e+17
## BIC 2.341e+03 2.312e+03 2.297e+03 2.281e+03 2.285e+03 2.293e+03 2.322e+03
The total assets of households react differently to OMO and MRO and
shocks. From half of the time periods, the reaction to OMO shocks is the
opposite, the reaction to MRO shocks is positive. Furthermore, we can
see in Figure … that the shock response of final consumption to CCI is
composed of different components. At the beginning of the period, the
shock response of marginal household consumption appears to be stronger
than at the end of the period. If the interaction between CCI and total
assets is taken into account, then the reaction to the third breakpoint
becomes negative, whereas it is marginally positive beforehand in the
reaction to CCI alone.
4.4 Credit channel
Anton (2015) described with the credit channel that the increase in
bank re serves and bank deposits due to expansionary MPs leads to an
increase in the volume of loans of households. The data for this channel
is available via the ECB database, therefore it is possible to examine
the significance of the credit channel. First, I examine the relation
between the MPs, MROs and non-reg ular OMOs, and the bank reserves and
the bank deposits. After this I examine the connection between the bank
reserves and the volume of loans. Anton (2015) sees the rising debt as a
crowding out effect for further indebtedness of private households
within his analysis for the years 2004 to 2014.19 How ever, Anton (2015)
analyzes this primarily based on the time series. My meth odology is
therefore based on the study “House prices and credit as transmis sion
channels from monetary policy to inequality: Evidence from OECD 19 Anton
(2015) summarizes this part of the analysis of the credit, cash flow and
liquidity channel. 41 countries” by Vale (2024). A panel vector
autoregression model was used here. Real GDP was used as a control
variable, as Vale (2024) uses the share of household credit in GDP from
bis-database as an endogenous variable. However, I used the adjusted
loans of households in the euro area.20
With a VAR model and an OLS regression model, it is possible to
investigate the transmission channel. The OLS can be used in several
phases. First, the link between the MPs of the ECB and the bank
reserves, then the link between the bank reserves and the quantity of
loans granted to households and finally the link between the loans of
households and the consumption of households can be examined. In the
next step, I will use the pvar model by Vale (2024) as a guide, as I
only have the aggregated eurozone data in this dataset and I will only
carry out a more country-specific differentiation in the next step, I
will first use a var model. In addition, an impulsive response function
can be used to analyze the effects of changes in the MPs of the ECB. For
the OLS regres sion, the three equations look like this: reservest = α +
β1 × MROt + β2 × OMOt + β3 × SovCISS t + ɛt, (11)21 bankloanst = α + β1
× reservest + β2 × realGDPt + β3 × unemploymentt + β4 × (MROt ×
reservest) + β5 × (OMOt × reservest) + ɛt, (12) Consumptiont = α + β1 ×
bankloanst + β2 × realGDPt + β3 × unemploymentt + β4 × (bankloanst ×
reservest) + ɛt. (13) The variable reserves from equations (11) and (12)
describe the total excess reserves of all credit institutions in the
euro area. The variable SovCISS de scribes the Sovereign Systemic Stress
Composite Indicator. I used it as an indicator for the stress of the
financial markets of the euro area, similar to the Financial Stress
Index (FSI) which is used by Mundra and Bicchal (2022) for their
analysis effects of monetary policy in India. The variable bankloans in
equations (12) and (13) represent the change in bank loans to
households. The consumption C in (13) is measured with the final
consumption. The variable 20 One reason for this is the available
frequency of the data. Vale (2024) analyzes a time horizon from 1995Q1
to 2019Q4 in her paper, so quarterly data is also appropriate. However,
here I need monthly data and I try to avoid approximating data,
especially since this is the endogenous variable. 21 This is monthly
data (2020M1 to 2024M11) taken from the ECB database. 42 realGDP in (12)
and (13) represent the realGDP growth of the euro area.22 In terms of
equation (11) the variable OMO is significant to the 0.01 level and the
variable MRO is significant to the 0.001 level.23 The coefficients of
MRO and OMO are as expected, MRO has a negative intercept and OMO a
positive intercept. In equation (12) all variables are significant, the
interaction term with OMO is only significant to 0.1 significance
level.24 Both interaction terms have a positive coefficient, the effect
of bank reserves and the effect of MRO and OMO influence each other
positively in terms of the effect on household loans. In equation (13)
only the control variable unemployment is not significant.25 The
interaction term has a positive coefficient, the effect of reserves and
household loans on household consumption influences each other
positively. For the further analysis, I use an autoregressive model, sim
ilar to the previous transmission channels, to take into account the
temporal shifts in the relationships. The variables are MRO, OMO,
SovCISS, bankloans, reserves, realGDP, un employment, bankdeposits26 and
final consumption. First, the respective time series of the data subset
are tested for stationarity using the augmented dickey-fuller test. None
of the variables is stationary at a critical value of 0.05. Now I test
the data subset for cointegration using the Johansen test. The Jo hansen
test suggests four cointegration relationships, in any case r ≥ 1. The
number of lags L are estimated as one. With the help of the regression
(11), the credit channel can also be proven to be largely significant.
Only the cor relation between bank deposits and household consumption is
weak. There fore, a VECM model is relatively evident. Without the
implication of interac tion terms in the VECM, almost exclusively the
last part of the transmission channel can be traced through significant
connections. Final consumption is, 22 The data for the variable realGDP
is available at the ECBs data base, the data is available in a quarterly
frequency, therefore the two months in between are estimated. 23 In the
regression to model (11), an R2 of 0.87. 24 For the sake of
completeness, OMO and MRO are of course directly included as variables
in the actual regression; these are not significant. These were not
taken into account in the model presentation, as they are not the focus
of the content. 25 Further results of the regressions are in appendix at
B.18 for (11), at
B.19 for (12) and at B.20 for (13). 26 Overnight deposits in the euro
area, excluding the central governments. 43 in addition to the control
variable unemployment, significantly negatively in fluenced by bank
reserves, MRO and hausholdloans. Under the implication of the
interaction terms, the short-term relationships change somewhat. Since
there are a total of three interaction terms with four affected
variables, the multicollinearity of several variables is unavoidable,
which complicates the estimation of an autoregressive model for this
transmission channel that takes the interaction terms into account. The
three interaction terms, MRO and OMO with reserves and with
householdloans, can be estimated with their co efficients using a VAR,
whereby, with the exception of the coefficients of these interaction
terms, these results are not discussed in more detail here, as important
prerequisites such as stationarity and no cointegration are not given
for the estimation of the VAR. The first part of the interaction terms
concerns the interaction between the MPs and the bank reserves. The
coefficient of the interaction of MRO with reserves is positive with
respect to household loans at each lag. The coefficient related to the
interaction term with OMO is rela tively low. The coefficient related to
final consumption of the interaction term regarding the interaction of
reserves and householdloans becomes positive with increasing lag,
i.e. the effect is initially negative related to final con sumption,
then becomes neutral and finally positive at the third lag. To further
track the interactions, the IRFs are analyzed to investigate the
structure of the effect.
##
## Call:
## lm(formula = dataeurozone$reserves ~ dataeurozone$MRO + dataeurozone$OMO +
## dataeurozone$SovCISS, data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1701660 -319603 28530 261631 1324028
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 666176.4 364062.7 1.830 0.0734 .
## dataeurozone$MRO -276313.1 80952.7 -3.413 0.0013 **
## dataeurozone$OMO 1321.5 202.3 6.532 3.52e-08 ***
## dataeurozone$SovCISS -1529849.7 662943.3 -2.308 0.0253 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 576000 on 49 degrees of freedom
## (7 observations deleted due to missingness)
## Multiple R-squared: 0.8705, Adjusted R-squared: 0.8626
## F-statistic: 109.8 on 3 and 49 DF, p-value: < 2.2e-16##
## Call:
## lm(formula = dataeurozone$Householdloans ~ dataeurozone$reserves +
## dataeurozone$GDP + dataeurozone$unemployment + (dataeurozone$reserves *
## dataeurozone$MRO) + (dataeurozone$reserves * dataeurozone$OMO),
## data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.93474 -0.19900 0.02601 0.16403 0.77123
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -5.869e+00 4.401e+00 -1.333 0.189082
## dataeurozone$reserves -1.644e-06 6.825e-07 -2.409 0.020152
## dataeurozone$GDP 1.048e+01 2.976e+00 3.522 0.000995
## dataeurozone$unemployment -4.269e-01 1.324e-01 -3.223 0.002359
## dataeurozone$MRO -1.041e+00 1.808e-01 -5.756 7.17e-07
## dataeurozone$OMO -4.878e-05 4.425e-04 -0.110 0.912708
## dataeurozone$reserves:dataeurozone$MRO 2.411e-07 9.134e-08 2.640 0.011357
## dataeurozone$reserves:dataeurozone$OMO 4.932e-10 2.820e-10 1.749 0.087170
##
## (Intercept)
## dataeurozone$reserves *
## dataeurozone$GDP ***
## dataeurozone$unemployment **
## dataeurozone$MRO ***
## dataeurozone$OMO
## dataeurozone$reserves:dataeurozone$MRO *
## dataeurozone$reserves:dataeurozone$OMO .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3221 on 45 degrees of freedom
## (7 observations deleted due to missingness)
## Multiple R-squared: 0.9555, Adjusted R-squared: 0.9486
## F-statistic: 138 on 7 and 45 DF, p-value: < 2.2e-16##
## Call:
## lm(formula = dataeurozone$finalconsumption ~ dataeurozone$Householdloans +
## dataeurozone$unemployment + dataeurozone$GDP + (dataeurozone$Householdloans *
## dataeurozone$reserves), data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -238772 -14525 723 35202 94573
##
## Coefficients:
## Estimate Std. Error t value
## (Intercept) 9.860e+05 5.904e+05 1.670
## dataeurozone$Householdloans -7.013e+04 1.492e+04 -4.701
## dataeurozone$unemployment -3.573e+04 3.337e+04 -1.071
## dataeurozone$GDP 9.469e+05 3.876e+05 2.443
## dataeurozone$reserves -3.590e-01 7.326e-02 -4.901
## dataeurozone$Householdloans:dataeurozone$reserves 7.674e-02 1.650e-02 4.650
## Pr(>|t|)
## (Intercept) 0.1010
## dataeurozone$Householdloans 2.01e-05 ***
## dataeurozone$unemployment 0.2893
## dataeurozone$GDP 0.0181 *
## dataeurozone$reserves 1.01e-05 ***
## dataeurozone$Householdloans:dataeurozone$reserves 2.39e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 61550 on 51 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.9028, Adjusted R-squared: 0.8933
## F-statistic: 94.74 on 5 and 51 DF, p-value: < 2.2e-16##
## Augmented Dickey-Fuller Test
##
## data: cOMO
## Dickey-Fuller = -2.4946, Lag order = 3, p-value = 0.3754
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: cMRO
## Dickey-Fuller = -2.4526, Lag order = 3, p-value = 0.3919
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: cTotalbankreserves
## Dickey-Fuller = -2.3355, Lag order = 3, p-value = 0.4392
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: cHouseholdloans
## Dickey-Fuller = -3.2619, Lag order = 3, p-value = 0.08632
## alternative hypothesis: stationary##
## Augmented Dickey-Fuller Test
##
## data: cFinalconsumption
## Dickey-Fuller = -2.211, Lag order = 3, p-value = 0.4895
## alternative hypothesis: stationary## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 9 9 9 9
##
## $criteria
## 1 2 3 4 5
## AIC(n) 4.449617e+01 4.357749e+01 4.373439e+01 4.299390e+01 4.238465e+01
## HQ(n) 4.494929e+01 4.440821e+01 4.494272e+01 4.457983e+01 4.434819e+01
## SC(n) 4.572491e+01 4.583019e+01 4.701104e+01 4.729450e+01 4.770921e+01
## FPE(n) 2.130202e+19 8.926789e+18 1.188683e+19 7.410122e+18 6.659718e+18
## 6 7 8 9 10
## AIC(n) 3.781290e+01 3.264569e+01 -7.500696e+01 -Inf -Inf
## HQ(n) 4.015404e+01 3.536443e+01 -7.191062e+01 -Inf -Inf
## SC(n) 4.416142e+01 4.001816e+01 -6.661054e+01 -Inf -Inf
## FPE(n) 1.742719e+17 6.375827e+15 2.512977e-29 0 0##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend in cointegration
##
## Eigenvalues (lambda):
## [1] 9.749071e-01 7.921490e-01 6.477729e-01 5.283576e-01 2.706390e-01
## [6] -4.996004e-16
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 4 | 14.83 10.49 12.25 16.26
## r <= 3 | 50.15 22.76 25.32 30.45
## r <= 2 | 99.20 39.06 42.44 48.45
## r <= 1 | 173.03 59.14 62.99 70.05
## r = 0 | 346.24 83.20 87.31 96.58
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## MRO.l1 OMO.l1 reserves.l1 Householdloans.l1
## MRO.l1 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## OMO.l1 -6.230827e-04 -1.633645e-03 -3.573917e-03 1.834638e-03
## reserves.l1 8.746218e-08 8.678386e-07 2.264156e-06 -2.439115e-07
## Householdloans.l1 9.597909e-01 5.687067e-01 2.570748e+00 3.828127e-03
## finalconsumption.l1 -8.630672e-07 3.609106e-06 -9.878148e-05 7.588448e-06
## trend.l1 -4.291277e-02 -1.301432e-01 1.199138e+00 -1.598240e-01
## finalconsumption.l1 trend.l1
## MRO.l1 1.000000e+00 1.000000e+00
## OMO.l1 2.316564e-03 3.624722e-03
## reserves.l1 4.018369e-07 -5.413492e-07
## Householdloans.l1 -6.864896e-01 -5.742840e-01
## finalconsumption.l1 1.120513e-05 3.329893e-05
## trend.l1 -1.714703e-01 -4.576525e-01
##
## Weights W:
## (This is the loading matrix)
##
## MRO.l1 OMO.l1 reserves.l1 Householdloans.l1
## MRO.d -1.410643e-01 -1.688733e-01 -5.608403e-03 -4.415793e-01
## OMO.d -3.294987e+02 3.951709e+02 2.008293e+01 6.680356e+01
## reserves.d 3.182352e+05 -1.059715e+06 5.499802e+04 1.286183e+06
## Householdloans.d -5.117124e-01 -4.497214e-02 3.694115e-02 1.253526e-01
## finalconsumption.d 4.750312e+04 2.543968e+03 4.090539e+03 -8.191049e+03
## finalconsumption.l1 trend.l1
## MRO.d -4.334208e-02 3.411511e-11
## OMO.d -6.696152e+01 -3.822832e-09
## reserves.d -4.813130e+05 4.004696e-05
## Householdloans.d 3.889161e-02 2.207763e-11
## finalconsumption.d -1.704904e+03 1.494508e-06## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 6 6 6 7
##
## $criteria
## 1 2 3 4 5 6
## AIC(n) 3.367692e+01 3.246247e+01 3.152359e+01 2.889503e+01 2.239628e+01 -Inf
## HQ(n) 3.452275e+01 3.404840e+01 3.384962e+01 3.196116e+01 2.620251e+01 -Inf
## SC(n) 3.597057e+01 3.676307e+01 3.783115e+01 3.720953e+01 3.271773e+01 -Inf
## FPE(n) 4.355403e+14 1.553058e+14 1.034239e+14 2.671996e+13 1.009135e+12 NaN
## 7 8 9 10
## AIC(n) -Inf -Inf -Inf -Inf
## HQ(n) -Inf -Inf -Inf -Inf
## SC(n) -Inf -Inf -Inf -Inf
## FPE(n) 0 0 0 0## Response MRO.d :
##
## Call:
## lm(formula = MRO.d ~ ect1 + ect2 + ect3 + ect4 + ect5 + ect6 +
## constant + MRO.dl1 + OMO.dl1 + SovCISS.dl1 + unemployment.dl1 +
## reserves.dl1 + Householdloans.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + SovCISS.dl2 + unemployment.dl2 + reserves.dl2 +
## Householdloans.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 +
## SovCISS.dl3 + unemployment.dl3 + reserves.dl3 + Householdloans.dl3 +
## finalconsumption.dl3 + MRO.dl4 + OMO.dl4 + SovCISS.dl4 +
## unemployment.dl4 + reserves.dl4 + Householdloans.dl4 + finalconsumption.dl4 -
## 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.12587 -0.02962 0.00074 0.02814 0.09488
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 -9.056e-02 3.379e-01 -0.268 0.79289
## ect2 -1.479e-03 6.813e-04 -2.171 0.04903 *
## ect3 2.008e+00 5.652e-01 3.554 0.00353 **
## ect4 5.205e-01 3.692e-01 1.410 0.18206
## ect5 5.276e-07 2.858e-07 1.846 0.08779 .
## ect6 3.246e-01 3.967e-01 0.818 0.42792
## constant -1.546e+00 6.456e+00 -0.240 0.81445
## MRO.dl1 -4.292e-01 3.696e-01 -1.161 0.26648
## OMO.dl1 1.342e-03 6.424e-04 2.089 0.05693 .
## SovCISS.dl1 -1.965e+00 1.284e+00 -1.530 0.14989
## unemployment.dl1 -4.706e-01 3.367e-01 -1.398 0.18559
## reserves.dl1 -4.965e-07 2.411e-07 -2.059 0.06009 .
## Householdloans.dl1 -6.080e-01 5.436e-01 -1.118 0.28362
## finalconsumption.dl1 3.340e-06 2.622e-06 1.274 0.22501
## MRO.dl2 -2.570e-01 4.886e-01 -0.526 0.60768
## OMO.dl2 8.585e-04 4.652e-04 1.846 0.08785 .
## SovCISS.dl2 -2.513e+00 1.571e+00 -1.600 0.13355
## unemployment.dl2 -5.653e-01 2.641e-01 -2.140 0.05186 .
## reserves.dl2 -5.392e-07 2.208e-07 -2.442 0.02965 *
## Householdloans.dl2 -5.663e-01 3.706e-01 -1.528 0.15039
## finalconsumption.dl2 4.451e-06 1.838e-06 2.422 0.03078 *
## MRO.dl3 -4.331e-02 4.556e-01 -0.095 0.92571
## OMO.dl3 1.482e-04 3.779e-04 0.392 0.70132
## SovCISS.dl3 -2.782e+00 1.241e+00 -2.241 0.04314 *
## unemployment.dl3 -3.638e-01 1.692e-01 -2.150 0.05097 .
## reserves.dl3 -1.304e-07 1.487e-07 -0.877 0.39657
## Householdloans.dl3 -5.466e-01 3.158e-01 -1.731 0.10712
## finalconsumption.dl3 1.327e-07 2.730e-06 0.049 0.96196
## MRO.dl4 -5.692e-02 3.005e-01 -0.189 0.85270
## OMO.dl4 2.988e-05 2.560e-04 0.117 0.90888
## SovCISS.dl4 -1.897e+00 1.183e+00 -1.604 0.13273
## unemployment.dl4 -2.064e-01 1.850e-01 -1.116 0.28477
## reserves.dl4 -3.248e-07 1.298e-07 -2.503 0.02646 *
## Householdloans.dl4 -4.503e-01 2.133e-01 -2.111 0.05468 .
## finalconsumption.dl4 7.991e-07 2.330e-06 0.343 0.73706
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1001 on 13 degrees of freedom
## Multiple R-squared: 0.9451, Adjusted R-squared: 0.7974
## F-statistic: 6.397 on 35 and 13 DF, p-value: 0.0004646
##
##
## Response OMO.d :
##
## Call:
## lm(formula = OMO.d ~ ect1 + ect2 + ect3 + ect4 + ect5 + ect6 +
## constant + MRO.dl1 + OMO.dl1 + SovCISS.dl1 + unemployment.dl1 +
## reserves.dl1 + Householdloans.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + SovCISS.dl2 + unemployment.dl2 + reserves.dl2 +
## Householdloans.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 +
## SovCISS.dl3 + unemployment.dl3 + reserves.dl3 + Householdloans.dl3 +
## finalconsumption.dl3 + MRO.dl4 + OMO.dl4 + SovCISS.dl4 +
## unemployment.dl4 + reserves.dl4 + Householdloans.dl4 + finalconsumption.dl4 -
## 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -117.802 -25.368 0.543 24.305 114.776
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 7.051e+02 3.111e+02 2.266 0.0412 *
## ect2 -7.618e-01 6.273e-01 -1.214 0.2462
## ect3 7.592e+02 5.204e+02 1.459 0.1683
## ect4 3.423e+02 3.399e+02 1.007 0.3323
## ect5 5.158e-04 2.632e-04 1.960 0.0718 .
## ect6 6.812e+02 3.653e+02 1.865 0.0849 .
## constant -7.580e+03 5.944e+03 -1.275 0.2245
## MRO.dl1 -9.591e+02 3.403e+02 -2.818 0.0145 *
## OMO.dl1 -1.741e-01 5.915e-01 -0.294 0.7732
## SovCISS.dl1 -3.512e+03 1.182e+03 -2.971 0.0108 *
## unemployment.dl1 -5.136e+02 3.100e+02 -1.657 0.1215
## reserves.dl1 -2.789e-04 2.220e-04 -1.256 0.2311
## Householdloans.dl1 -6.557e+02 5.005e+02 -1.310 0.2128
## finalconsumption.dl1 -4.589e-03 2.414e-03 -1.901 0.0797 .
## MRO.dl2 -9.864e+02 4.498e+02 -2.193 0.0471 *
## OMO.dl2 -4.205e-01 4.283e-01 -0.982 0.3441
## SovCISS.dl2 -2.983e+03 1.446e+03 -2.063 0.0597 .
## unemployment.dl2 9.812e+00 2.432e+02 0.040 0.9684
## reserves.dl2 -1.868e-04 2.033e-04 -0.919 0.3748
## Householdloans.dl2 -3.756e+02 3.412e+02 -1.101 0.2909
## finalconsumption.dl2 -3.316e-03 1.692e-03 -1.960 0.0718 .
## MRO.dl3 -1.181e+03 4.194e+02 -2.816 0.0146 *
## OMO.dl3 -3.665e-01 3.480e-01 -1.053 0.3114
## SovCISS.dl3 -1.776e+03 1.143e+03 -1.553 0.1443
## unemployment.dl3 5.443e+01 1.558e+02 0.349 0.7324
## reserves.dl3 -1.410e-04 1.369e-04 -1.030 0.3220
## Householdloans.dl3 -7.592e+01 2.908e+02 -0.261 0.7981
## finalconsumption.dl3 -3.677e-03 2.513e-03 -1.463 0.1672
## MRO.dl4 -6.308e+02 2.767e+02 -2.280 0.0401 *
## OMO.dl4 -8.602e-02 2.357e-01 -0.365 0.7210
## SovCISS.dl4 -1.008e+03 1.089e+03 -0.925 0.3716
## unemployment.dl4 3.136e+02 1.703e+02 1.842 0.0885 .
## reserves.dl4 3.522e-05 1.195e-04 0.295 0.7728
## Householdloans.dl4 4.445e+01 1.964e+02 0.226 0.8245
## finalconsumption.dl4 -3.459e-03 2.145e-03 -1.613 0.1308
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 92.19 on 13 degrees of freedom
## Multiple R-squared: 0.9054, Adjusted R-squared: 0.6508
## F-statistic: 3.556 on 35 and 13 DF, p-value: 0.008832
##
##
## Response SovCISS.d :
##
## Call:
## lm(formula = SovCISS.d ~ ect1 + ect2 + ect3 + ect4 + ect5 + ect6 +
## constant + MRO.dl1 + OMO.dl1 + SovCISS.dl1 + unemployment.dl1 +
## reserves.dl1 + Householdloans.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + SovCISS.dl2 + unemployment.dl2 + reserves.dl2 +
## Householdloans.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 +
## SovCISS.dl3 + unemployment.dl3 + reserves.dl3 + Householdloans.dl3 +
## finalconsumption.dl3 + MRO.dl4 + OMO.dl4 + SovCISS.dl4 +
## unemployment.dl4 + reserves.dl4 + Householdloans.dl4 + finalconsumption.dl4 -
## 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.036800 -0.007447 -0.000788 0.007232 0.048320
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 6.513e-02 9.960e-02 0.654 0.525
## ect2 6.505e-05 2.008e-04 0.324 0.751
## ect3 -3.461e-02 1.666e-01 -0.208 0.839
## ect4 -8.769e-03 1.088e-01 -0.081 0.937
## ect5 -4.073e-09 8.424e-08 -0.048 0.962
## ect6 4.168e-02 1.169e-01 0.356 0.727
## constant 5.555e-02 1.903e+00 0.029 0.977
## MRO.dl1 -1.135e-01 1.090e-01 -1.042 0.316
## OMO.dl1 -1.572e-04 1.894e-04 -0.830 0.421
## SovCISS.dl1 1.908e-01 3.784e-01 0.504 0.623
## unemployment.dl1 4.384e-02 9.925e-02 0.442 0.666
## reserves.dl1 2.485e-09 7.107e-08 0.035 0.973
## Householdloans.dl1 -9.729e-02 1.602e-01 -0.607 0.554
## finalconsumption.dl1 -1.305e-07 7.729e-07 -0.169 0.869
## MRO.dl2 -1.270e-01 1.440e-01 -0.882 0.394
## OMO.dl2 -1.178e-04 1.371e-04 -0.859 0.406
## SovCISS.dl2 -2.661e-01 4.629e-01 -0.575 0.575
## unemployment.dl2 3.493e-02 7.785e-02 0.449 0.661
## reserves.dl2 2.324e-08 6.508e-08 0.357 0.727
## Householdloans.dl2 8.695e-02 1.092e-01 0.796 0.440
## finalconsumption.dl2 -4.281e-07 5.417e-07 -0.790 0.443
## MRO.dl3 -1.654e-02 1.343e-01 -0.123 0.904
## OMO.dl3 -5.871e-05 1.114e-04 -0.527 0.607
## SovCISS.dl3 4.906e-01 3.659e-01 1.341 0.203
## unemployment.dl3 9.401e-02 4.988e-02 1.885 0.082 .
## reserves.dl3 4.173e-08 4.384e-08 0.952 0.359
## Householdloans.dl3 5.337e-04 9.308e-02 0.006 0.996
## finalconsumption.dl3 -1.756e-07 8.046e-07 -0.218 0.831
## MRO.dl4 -3.607e-02 8.858e-02 -0.407 0.690
## OMO.dl4 7.520e-06 7.546e-05 0.100 0.922
## SovCISS.dl4 1.441e-01 3.486e-01 0.413 0.686
## unemployment.dl4 2.529e-02 5.452e-02 0.464 0.650
## reserves.dl4 7.048e-09 3.826e-08 0.184 0.857
## Householdloans.dl4 4.341e-02 6.287e-02 0.691 0.502
## finalconsumption.dl4 2.917e-07 6.867e-07 0.425 0.678
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.02951 on 13 degrees of freedom
## Multiple R-squared: 0.7522, Adjusted R-squared: 0.08508
## F-statistic: 1.128 on 35 and 13 DF, p-value: 0.4265
##
##
## Response unemployment.d :
##
## Call:
## lm(formula = unemployment.d ~ ect1 + ect2 + ect3 + ect4 + ect5 +
## ect6 + constant + MRO.dl1 + OMO.dl1 + SovCISS.dl1 + unemployment.dl1 +
## reserves.dl1 + Householdloans.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + SovCISS.dl2 + unemployment.dl2 + reserves.dl2 +
## Householdloans.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 +
## SovCISS.dl3 + unemployment.dl3 + reserves.dl3 + Householdloans.dl3 +
## finalconsumption.dl3 + MRO.dl4 + OMO.dl4 + SovCISS.dl4 +
## unemployment.dl4 + reserves.dl4 + Householdloans.dl4 + finalconsumption.dl4 -
## 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.169945 -0.038548 0.001429 0.040015 0.168856
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 -7.457e-01 3.862e-01 -1.931 0.0756 .
## ect2 9.570e-04 7.787e-04 1.229 0.2408
## ect3 9.319e-01 6.459e-01 1.443 0.1728
## ect4 -7.609e-01 4.220e-01 -1.803 0.0946 .
## ect5 -3.188e-07 3.266e-07 -0.976 0.3469
## ect6 -9.400e-01 4.534e-01 -2.073 0.0586 .
## constant 1.605e-01 7.378e+00 0.022 0.9830
## MRO.dl1 7.134e-01 4.224e-01 1.689 0.1151
## OMO.dl1 -5.471e-04 7.342e-04 -0.745 0.4694
## SovCISS.dl1 -2.813e+00 1.467e+00 -1.917 0.0774 .
## unemployment.dl1 1.939e-01 3.848e-01 0.504 0.6228
## reserves.dl1 2.309e-07 2.756e-07 0.838 0.4171
## Householdloans.dl1 1.252e+00 6.213e-01 2.016 0.0649 .
## finalconsumption.dl1 -5.874e-07 2.997e-06 -0.196 0.8476
## MRO.dl2 4.197e-01 5.584e-01 0.752 0.4656
## OMO.dl2 -1.129e-03 5.316e-04 -2.123 0.0535 .
## SovCISS.dl2 -1.193e+00 1.795e+00 -0.664 0.5181
## unemployment.dl2 -9.036e-02 3.018e-01 -0.299 0.7694
## reserves.dl2 1.260e-07 2.523e-07 0.499 0.6260
## Householdloans.dl2 3.320e-01 4.235e-01 0.784 0.4471
## finalconsumption.dl2 -4.906e-06 2.100e-06 -2.336 0.0362 *
## MRO.dl3 1.564e-01 5.207e-01 0.300 0.7686
## OMO.dl3 -3.347e-04 4.319e-04 -0.775 0.4523
## SovCISS.dl3 -3.553e+00 1.419e+00 -2.504 0.0264 *
## unemployment.dl3 -5.773e-02 1.934e-01 -0.298 0.7700
## reserves.dl3 8.932e-08 1.700e-07 0.525 0.6081
## Householdloans.dl3 1.678e-01 3.609e-01 0.465 0.6496
## finalconsumption.dl3 -5.680e-06 3.120e-06 -1.821 0.0917 .
## MRO.dl4 4.683e-01 3.434e-01 1.364 0.1958
## OMO.dl4 -2.524e-04 2.926e-04 -0.863 0.4039
## SovCISS.dl4 -8.113e-01 1.352e+00 -0.600 0.5587
## unemployment.dl4 -1.944e-01 2.114e-01 -0.920 0.3745
## reserves.dl4 1.451e-07 1.484e-07 0.978 0.3457
## Householdloans.dl4 2.541e-01 2.438e-01 1.043 0.3162
## finalconsumption.dl4 3.634e-07 2.662e-06 0.137 0.8935
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1144 on 13 degrees of freedom
## Multiple R-squared: 0.9263, Adjusted R-squared: 0.7279
## F-statistic: 4.669 on 35 and 13 DF, p-value: 0.002388
##
##
## Response reserves.d :
##
## Call:
## lm(formula = reserves.d ~ ect1 + ect2 + ect3 + ect4 + ect5 +
## ect6 + constant + MRO.dl1 + OMO.dl1 + SovCISS.dl1 + unemployment.dl1 +
## reserves.dl1 + Householdloans.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + SovCISS.dl2 + unemployment.dl2 + reserves.dl2 +
## Householdloans.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 +
## SovCISS.dl3 + unemployment.dl3 + reserves.dl3 + Householdloans.dl3 +
## finalconsumption.dl3 + MRO.dl4 + OMO.dl4 + SovCISS.dl4 +
## unemployment.dl4 + reserves.dl4 + Householdloans.dl4 + finalconsumption.dl4 -
## 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -304834 -115566 2351 69590 378925
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 -2.405e+06 9.642e+05 -2.495 0.02686 *
## ect2 6.608e+03 1.944e+03 3.399 0.00475 **
## ect3 1.058e+06 1.613e+06 0.656 0.52334
## ect4 -2.039e+06 1.054e+06 -1.935 0.07502 .
## ect5 -2.628e+00 8.156e-01 -3.223 0.00667 **
## ect6 -3.009e+06 1.132e+06 -2.658 0.01972 *
## constant 8.504e+06 1.842e+07 0.462 0.65200
## MRO.dl1 2.070e+06 1.055e+06 1.962 0.07151 .
## OMO.dl1 -4.833e+03 1.833e+03 -2.637 0.02053 *
## SovCISS.dl1 -2.087e+05 3.664e+06 -0.057 0.95544
## unemployment.dl1 -1.197e+05 9.608e+05 -0.125 0.90278
## reserves.dl1 1.798e+00 6.881e-01 2.613 0.02148 *
## Householdloans.dl1 2.616e+06 1.551e+06 1.686 0.11555
## finalconsumption.dl1 -1.189e+01 7.483e+00 -1.589 0.13607
## MRO.dl2 1.627e+06 1.394e+06 1.167 0.26430
## OMO.dl2 -2.199e+03 1.327e+03 -1.657 0.12147
## SovCISS.dl2 -8.123e+06 4.482e+06 -1.813 0.09305 .
## unemployment.dl2 4.311e+05 7.537e+05 0.572 0.57706
## reserves.dl2 1.378e+00 6.300e-01 2.187 0.04766 *
## Householdloans.dl2 2.146e+06 1.057e+06 2.029 0.06345 .
## finalconsumption.dl2 -1.623e+00 5.244e+00 -0.310 0.76180
## MRO.dl3 -7.337e+05 1.300e+06 -0.564 0.58213
## OMO.dl3 -1.923e+03 1.078e+03 -1.784 0.09785 .
## SovCISS.dl3 1.785e+06 3.543e+06 0.504 0.62271
## unemployment.dl3 5.980e+04 4.829e+05 0.124 0.90335
## reserves.dl3 5.784e-01 4.245e-01 1.363 0.19611
## Householdloans.dl3 -1.392e+05 9.012e+05 -0.154 0.87962
## finalconsumption.dl3 -6.057e+00 7.790e+00 -0.778 0.45073
## MRO.dl4 -2.282e+06 8.576e+05 -2.661 0.01959 *
## OMO.dl4 -6.100e+02 7.306e+02 -0.835 0.41886
## SovCISS.dl4 -4.135e+06 3.375e+06 -1.225 0.24226
## unemployment.dl4 -5.450e+05 5.279e+05 -1.033 0.32064
## reserves.dl4 -7.933e-02 3.704e-01 -0.214 0.83373
## Householdloans.dl4 4.829e+05 6.087e+05 0.793 0.44184
## finalconsumption.dl4 -1.106e+01 6.648e+00 -1.663 0.12015
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 285700 on 13 degrees of freedom
## Multiple R-squared: 0.8807, Adjusted R-squared: 0.5596
## F-statistic: 2.742 on 35 and 13 DF, p-value: 0.02744
##
##
## Response Householdloans.d :
##
## Call:
## lm(formula = Householdloans.d ~ ect1 + ect2 + ect3 + ect4 + ect5 +
## ect6 + constant + MRO.dl1 + OMO.dl1 + SovCISS.dl1 + unemployment.dl1 +
## reserves.dl1 + Householdloans.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + SovCISS.dl2 + unemployment.dl2 + reserves.dl2 +
## Householdloans.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 +
## SovCISS.dl3 + unemployment.dl3 + reserves.dl3 + Householdloans.dl3 +
## finalconsumption.dl3 + MRO.dl4 + OMO.dl4 + SovCISS.dl4 +
## unemployment.dl4 + reserves.dl4 + Householdloans.dl4 + finalconsumption.dl4 -
## 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.086615 -0.020432 0.000039 0.020796 0.094096
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 1.530e-01 2.595e-01 0.590 0.5656
## ect2 2.535e-04 5.233e-04 0.484 0.6362
## ect3 4.243e-01 4.341e-01 0.978 0.3462
## ect4 8.037e-03 2.836e-01 0.028 0.9778
## ect5 2.132e-07 2.195e-07 0.971 0.3492
## ect6 -5.685e-02 3.047e-01 -0.187 0.8549
## constant 1.108e+00 4.958e+00 0.223 0.8267
## MRO.dl1 -2.969e-01 2.839e-01 -1.046 0.3147
## OMO.dl1 -1.180e-04 4.934e-04 -0.239 0.8147
## SovCISS.dl1 -1.262e+00 9.861e-01 -1.280 0.2229
## unemployment.dl1 -2.866e-01 2.586e-01 -1.108 0.2878
## reserves.dl1 -1.816e-07 1.852e-07 -0.981 0.3446
## Householdloans.dl1 -2.896e-01 4.175e-01 -0.694 0.5001
## finalconsumption.dl1 -1.142e-06 2.014e-06 -0.567 0.5803
## MRO.dl2 -4.561e-01 3.752e-01 -1.216 0.2458
## OMO.dl2 -4.584e-04 3.573e-04 -1.283 0.2219
## SovCISS.dl2 -1.395e+00 1.206e+00 -1.156 0.2683
## unemployment.dl2 1.313e-01 2.029e-01 0.647 0.5286
## reserves.dl2 -1.305e-07 1.696e-07 -0.770 0.4553
## Householdloans.dl2 -1.177e-01 2.846e-01 -0.413 0.6860
## finalconsumption.dl2 -2.560e-06 1.411e-06 -1.814 0.0928 .
## MRO.dl3 -6.491e-01 3.499e-01 -1.855 0.0864 .
## OMO.dl3 -1.403e-04 2.903e-04 -0.483 0.6369
## SovCISS.dl3 -1.585e-02 9.535e-01 -0.017 0.9870
## unemployment.dl3 1.791e-01 1.300e-01 1.378 0.1915
## reserves.dl3 -1.291e-07 1.142e-07 -1.130 0.2788
## Householdloans.dl3 -9.068e-02 2.426e-01 -0.374 0.7145
## finalconsumption.dl3 -1.288e-07 2.097e-06 -0.061 0.9519
## MRO.dl4 -2.313e-01 2.308e-01 -1.002 0.3346
## OMO.dl4 -1.204e-04 1.966e-04 -0.612 0.5510
## SovCISS.dl4 -1.265e+00 9.083e-01 -1.393 0.1871
## unemployment.dl4 2.110e-01 1.421e-01 1.485 0.1614
## reserves.dl4 -5.165e-08 9.970e-08 -0.518 0.6131
## Householdloans.dl4 1.108e-01 1.638e-01 0.676 0.5107
## finalconsumption.dl4 -3.255e-06 1.789e-06 -1.819 0.0920 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.0769 on 13 degrees of freedom
## Multiple R-squared: 0.9575, Adjusted R-squared: 0.8432
## F-statistic: 8.373 on 35 and 13 DF, p-value: 0.0001055
##
##
## Response finalconsumption.d :
##
## Call:
## lm(formula = finalconsumption.d ~ ect1 + ect2 + ect3 + ect4 +
## ect5 + ect6 + constant + MRO.dl1 + OMO.dl1 + SovCISS.dl1 +
## unemployment.dl1 + reserves.dl1 + Householdloans.dl1 + finalconsumption.dl1 +
## MRO.dl2 + OMO.dl2 + SovCISS.dl2 + unemployment.dl2 + reserves.dl2 +
## Householdloans.dl2 + finalconsumption.dl2 + MRO.dl3 + OMO.dl3 +
## SovCISS.dl3 + unemployment.dl3 + reserves.dl3 + Householdloans.dl3 +
## finalconsumption.dl3 + MRO.dl4 + OMO.dl4 + SovCISS.dl4 +
## unemployment.dl4 + reserves.dl4 + Householdloans.dl4 + finalconsumption.dl4 -
## 1, data = data.mat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -5055.1 -1408.7 -281.2 1247.9 5164.4
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## ect1 -5.727e+04 1.423e+04 -4.025 0.001442 **
## ect2 1.083e+02 2.869e+01 3.776 0.002312 **
## ect3 -6.425e+03 2.380e+04 -0.270 0.791411
## ect4 -9.499e+04 1.555e+04 -6.110 3.72e-05 ***
## ect5 -6.549e-02 1.204e-02 -5.441 0.000113 ***
## ect6 -5.014e+04 1.671e+04 -3.001 0.010217 *
## constant 2.591e+06 2.718e+05 9.532 3.13e-07 ***
## MRO.dl1 6.335e+04 1.556e+04 4.070 0.001325 **
## OMO.dl1 -6.515e+01 2.705e+01 -2.408 0.031595 *
## SovCISS.dl1 9.844e+04 5.406e+04 1.821 0.091711 .
## unemployment.dl1 1.045e+05 1.418e+04 7.371 5.42e-06 ***
## reserves.dl1 5.410e-02 1.015e-02 5.328 0.000137 ***
## Householdloans.dl1 6.580e+04 2.289e+04 2.874 0.013031 *
## finalconsumption.dl1 7.461e-01 1.104e-01 6.757 1.35e-05 ***
## MRO.dl2 7.922e+04 2.057e+04 3.851 0.002005 **
## OMO.dl2 -4.783e+01 1.959e+01 -2.442 0.029666 *
## SovCISS.dl2 1.873e+05 6.613e+04 2.832 0.014141 *
## unemployment.dl2 6.072e+04 1.112e+04 5.460 0.000109 ***
## reserves.dl2 4.432e-02 9.297e-03 4.767 0.000368 ***
## Householdloans.dl2 4.232e+04 1.560e+04 2.712 0.017776 *
## finalconsumption.dl2 4.770e-01 7.738e-02 6.164 3.41e-05 ***
## MRO.dl3 8.835e+04 1.918e+04 4.606 0.000493 ***
## OMO.dl3 1.740e+01 1.591e+01 1.093 0.294112
## SovCISS.dl3 5.628e+04 5.228e+04 1.077 0.301257
## unemployment.dl3 3.726e+04 7.126e+03 5.229 0.000163 ***
## reserves.dl3 3.317e-02 6.264e-03 5.296 0.000145 ***
## Householdloans.dl3 3.601e+04 1.330e+04 2.708 0.017933 *
## finalconsumption.dl3 3.304e-01 1.149e-01 2.874 0.013044 *
## MRO.dl4 5.769e+04 1.265e+04 4.559 0.000537 ***
## OMO.dl4 -8.581e+00 1.078e+01 -0.796 0.440350
## SovCISS.dl4 9.938e+04 4.980e+04 1.996 0.067376 .
## unemployment.dl4 4.439e+03 7.789e+03 0.570 0.578482
## reserves.dl4 1.455e-02 5.466e-03 2.662 0.019565 *
## Householdloans.dl4 1.898e+04 8.982e+03 2.113 0.054529 .
## finalconsumption.dl4 5.122e-01 9.810e-02 5.221 0.000165 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4216 on 13 degrees of freedom
## Multiple R-squared: 0.9903, Adjusted R-squared: 0.9641
## F-statistic: 37.82 on 35 and 13 DF, p-value: 1.171e-08## ect1 ect2 ect3
## Min. :-1.4489456 Min. : -0.0392 Min. :-0.1405110
## 1st Qu.: 0.0000000 1st Qu.: 0.0000 1st Qu.: 0.0000000
## Median : 0.0000000 Median : 0.0000 Median : 0.0000000
## Mean :-0.0561045 Mean : 62.9572 Mean : 0.1074375
## 3rd Qu.: 0.0000274 3rd Qu.: 0.2500 3rd Qu.: 0.0000027
## Max. : 1.0000000 Max. :502.6964 Max. : 1.0000000
## ect4 ect5 ect6
## Min. :-2.0e-01 Min. : -89.4 Min. :-0.0001086
## 1st Qu.: 0.0e+00 1st Qu.: 0.0 1st Qu.: 0.0000000
## Median : 0.0e+00 Median : 0.0 Median : 0.0000000
## Mean : 1.0e-01 Mean : 142044.3 Mean : 0.2979746
## 3rd Qu.: 5.1e-06 3rd Qu.: 0.2 3rd Qu.: 0.2500000
## Max. : 1.0e+00 Max. :1136442.7 Max. : 1.3839054## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 3 3 3 2
##
## $criteria
## 1 2 3 4 5 6 7 8 9 10
## AIC(n) -5.41101628 NaN -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## HQ(n) -5.13818741 NaN -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## SC(n) -2.66631350 NaN -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## FPE(n) 0.00784646 -1.125729e-77 0 0 0 0 0 0 0 0## [1] 7## [1] 7## MRO OMO SovCISS unemployment reserves
## MRO 1.0000000 -0.9694646 -0.6971271 -0.3553923 -0.9060649
## OMO -0.9694646 1.0000000 0.8108057 0.3644879 0.8462709
## SovCISS -0.6971271 0.8108057 1.0000000 0.1972952 0.4922168
## unemployment -0.3553923 0.3644879 0.1972952 1.0000000 0.2144227
## reserves -0.9060649 0.8462709 0.4922168 0.2144227 1.0000000
## Householdloans -0.9418832 0.9741303 0.8697506 0.3833306 0.7594852
## durablehs 0.7201660 -0.7949001 -0.8996729 -0.2800758 -0.5313504
## Householdloans durablehs
## MRO -0.9418832 0.7201660
## OMO 0.9741303 -0.7949001
## SovCISS 0.8697506 -0.8996729
## unemployment 0.3833306 -0.2800758
## reserves 0.7594852 -0.5313504
## Householdloans 1.0000000 -0.8493747
## durablehs -0.8493747 1.0000000## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 5 5 5 5
##
## $criteria
## 1 2 3 4 5 6 7 8
## AIC(n) 1.184072e+02 1.139654e+02 1.038211e+02 -1.823512e+02 -Inf -Inf -Inf -Inf
## HQ(n) 1.200686e+02 1.171373e+02 1.085034e+02 -1.761585e+02 -Inf -Inf -Inf -Inf
## SC(n) 1.229126e+02 1.225666e+02 1.165181e+02 -1.655584e+02 -Inf -Inf -Inf -Inf
## FPE(n) 2.978906e+51 7.765794e+49 5.339688e+46 5.707828e-72 0 0 0 0
## 9 10
## AIC(n) -Inf -Inf
## HQ(n) -Inf -Inf
## SC(n) -Inf -Inf
## FPE(n) 0 0##
## VAR Estimation Results:
## =======================
##
## Estimated coefficients for equation MRO:
## ========================================
## Call:
## MRO = MRO.l1 + OMO.l1 + reserves.l1 + Householdloans.l1 + SovCISS.l1 + unemployment.l1 + finalconsumption.l1 + interactionOMOreserves.l1 + interactionMROreserves.l1 + interactioncreditreservesloans.l1 + MRO.l2 + OMO.l2 + reserves.l2 + Householdloans.l2 + SovCISS.l2 + unemployment.l2 + finalconsumption.l2 + interactionOMOreserves.l2 + interactionMROreserves.l2 + interactioncreditreservesloans.l2 + MRO.l3 + OMO.l3 + reserves.l3 + Householdloans.l3 + SovCISS.l3 + unemployment.l3 + finalconsumption.l3 + interactionOMOreserves.l3 + interactionMROreserves.l3 + interactioncreditreservesloans.l3 + const
##
## MRO.l1 OMO.l1
## 1.091204e+00 -2.611277e-04
## reserves.l1 Householdloans.l1
## 1.474542e-06 5.735326e-01
## SovCISS.l1 unemployment.l1
## -1.631003e+00 -7.068903e-02
## finalconsumption.l1 interactionOMOreserves.l1
## 6.560903e-07 -3.116661e-07
## interactionMROreserves.l1 interactioncreditreservesloans.l1
## 1.557921e-10 -3.157045e-07
## MRO.l2 OMO.l2
## -3.690988e-01 -6.262471e-04
## reserves.l2 Householdloans.l2
## -8.494386e-07 3.082801e-02
## SovCISS.l2 unemployment.l2
## 1.713108e+00 1.183063e-01
## finalconsumption.l2 interactionOMOreserves.l2
## 5.341483e-07 3.902547e-07
## interactionMROreserves.l2 interactioncreditreservesloans.l2
## 1.130108e-10 9.797363e-08
## MRO.l3 OMO.l3
## 6.212481e-01 4.942052e-04
## reserves.l3 Householdloans.l3
## 4.712551e-07 -1.576228e-01
## SovCISS.l3 unemployment.l3
## 9.390983e-01 -6.506807e-02
## finalconsumption.l3 interactionOMOreserves.l3
## -5.446687e-07 -6.984004e-08
## interactionMROreserves.l3 interactioncreditreservesloans.l3
## -3.996595e-10 8.823921e-08
## const
## -2.925080e+00
##
##
## Estimated coefficients for equation OMO:
## ========================================
## Call:
## OMO = MRO.l1 + OMO.l1 + reserves.l1 + Householdloans.l1 + SovCISS.l1 + unemployment.l1 + finalconsumption.l1 + interactionOMOreserves.l1 + interactionMROreserves.l1 + interactioncreditreservesloans.l1 + MRO.l2 + OMO.l2 + reserves.l2 + Householdloans.l2 + SovCISS.l2 + unemployment.l2 + finalconsumption.l2 + interactionOMOreserves.l2 + interactionMROreserves.l2 + interactioncreditreservesloans.l2 + MRO.l3 + OMO.l3 + reserves.l3 + Householdloans.l3 + SovCISS.l3 + unemployment.l3 + finalconsumption.l3 + interactionOMOreserves.l3 + interactionMROreserves.l3 + interactioncreditreservesloans.l3 + const
##
## MRO.l1 OMO.l1
## -4.171794e+02 7.986175e-02
## reserves.l1 Householdloans.l1
## -3.568207e-04 -2.884112e+01
## SovCISS.l1 unemployment.l1
## 2.696271e+02 3.349529e+01
## finalconsumption.l1 interactionOMOreserves.l1
## -1.328820e-03 1.025345e-04
## interactionMROreserves.l1 interactioncreditreservesloans.l1
## -6.482568e-08 1.233900e-04
## MRO.l2 OMO.l2
## 6.333667e+02 -1.260807e-01
## reserves.l2 Householdloans.l2
## 1.101561e-03 2.109245e+02
## SovCISS.l2 unemployment.l2
## -9.632697e+01 8.661739e+01
## finalconsumption.l2 interactionOMOreserves.l2
## -1.061344e-03 -1.694971e-04
## interactionMROreserves.l2 interactioncreditreservesloans.l2
## 2.044165e-07 -3.047237e-04
## MRO.l3 OMO.l3
## -3.778978e+02 -4.657969e-01
## reserves.l3 Householdloans.l3
## -1.425685e-03 7.486362e+00
## SovCISS.l3 unemployment.l3
## 1.324321e+03 3.772195e+01
## finalconsumption.l3 interactionOMOreserves.l3
## 1.038962e-03 1.003061e-04
## interactionMROreserves.l3 interactioncreditreservesloans.l3
## 5.458232e-07 5.108755e-05
## const
## 2.521451e+03
##
##
## Estimated coefficients for equation reserves:
## =============================================
## Call:
## reserves = MRO.l1 + OMO.l1 + reserves.l1 + Householdloans.l1 + SovCISS.l1 + unemployment.l1 + finalconsumption.l1 + interactionOMOreserves.l1 + interactionMROreserves.l1 + interactioncreditreservesloans.l1 + MRO.l2 + OMO.l2 + reserves.l2 + Householdloans.l2 + SovCISS.l2 + unemployment.l2 + finalconsumption.l2 + interactionOMOreserves.l2 + interactionMROreserves.l2 + interactioncreditreservesloans.l2 + MRO.l3 + OMO.l3 + reserves.l3 + Householdloans.l3 + SovCISS.l3 + unemployment.l3 + finalconsumption.l3 + interactionOMOreserves.l3 + interactionMROreserves.l3 + interactioncreditreservesloans.l3 + const
##
## MRO.l1 OMO.l1
## 1.423259e+06 -9.175948e+01
## reserves.l1 Householdloans.l1
## 3.380069e+00 1.720921e+06
## SovCISS.l1 unemployment.l1
## -2.744785e+06 -5.699052e+05
## finalconsumption.l1 interactionOMOreserves.l1
## 1.194955e+00 -3.844194e-01
## interactionMROreserves.l1 interactioncreditreservesloans.l1
## 3.772056e-05 -7.000619e-01
## MRO.l2 OMO.l2
## -6.610102e+05 -5.424986e+02
## reserves.l2 Householdloans.l2
## -4.216832e-01 -2.661995e+05
## SovCISS.l2 unemployment.l2
## 1.581827e+06 1.505221e+05
## finalconsumption.l2 interactionOMOreserves.l2
## -1.234183e-01 -4.250779e-01
## interactionMROreserves.l2 interactioncreditreservesloans.l2
## -4.584764e-05 2.025725e-01
## MRO.l3 OMO.l3
## 2.912083e+05 1.132954e+03
## reserves.l3 Householdloans.l3
## 1.622354e+00 -9.306179e+05
## SovCISS.l3 unemployment.l3
## 9.816405e+05 -2.530435e+05
## finalconsumption.l3 interactionOMOreserves.l3
## -2.799981e+00 -6.874811e-01
## interactionMROreserves.l3 interactioncreditreservesloans.l3
## -1.092979e-03 3.407608e-01
## const
## 2.799841e+06
##
##
## Estimated coefficients for equation Householdloans:
## ===================================================
## Call:
## Householdloans = MRO.l1 + OMO.l1 + reserves.l1 + Householdloans.l1 + SovCISS.l1 + unemployment.l1 + finalconsumption.l1 + interactionOMOreserves.l1 + interactionMROreserves.l1 + interactioncreditreservesloans.l1 + MRO.l2 + OMO.l2 + reserves.l2 + Householdloans.l2 + SovCISS.l2 + unemployment.l2 + finalconsumption.l2 + interactionOMOreserves.l2 + interactionMROreserves.l2 + interactioncreditreservesloans.l2 + MRO.l3 + OMO.l3 + reserves.l3 + Householdloans.l3 + SovCISS.l3 + unemployment.l3 + finalconsumption.l3 + interactionOMOreserves.l3 + interactionMROreserves.l3 + interactioncreditreservesloans.l3 + const
##
## MRO.l1 OMO.l1
## -5.852919e-01 -7.165781e-05
## reserves.l1 Householdloans.l1
## -2.289057e-06 -2.883322e-02
## SovCISS.l1 unemployment.l1
## 2.097648e-02 -5.939372e-02
## finalconsumption.l1 interactionOMOreserves.l1
## 2.235954e-06 1.664282e-07
## interactionMROreserves.l1 interactioncreditreservesloans.l1
## 3.042255e-10 3.343842e-07
## MRO.l2 OMO.l2
## -2.544154e-02 -4.906958e-05
## reserves.l2 Householdloans.l2
## 7.051550e-07 -9.907610e-02
## SovCISS.l2 unemployment.l2
## 1.328690e+00 3.688917e-01
## finalconsumption.l2 interactionOMOreserves.l2
## -4.622956e-06 -3.334712e-09
## interactionMROreserves.l2 interactioncreditreservesloans.l2
## 3.658692e-11 -1.890162e-07
## MRO.l3 OMO.l3
## -5.244721e-01 -3.616576e-04
## reserves.l3 Householdloans.l3
## -7.908822e-07 3.885657e-01
## SovCISS.l3 unemployment.l3
## -4.302868e-01 -8.118398e-02
## finalconsumption.l3 interactionOMOreserves.l3
## 3.858979e-06 7.014214e-08
## interactionMROreserves.l3 interactioncreditreservesloans.l3
## 5.962373e-10 -1.292763e-07
## const
## 8.993693e-01
##
##
## Estimated coefficients for equation SovCISS:
## ============================================
## Call:
## SovCISS = MRO.l1 + OMO.l1 + reserves.l1 + Householdloans.l1 + SovCISS.l1 + unemployment.l1 + finalconsumption.l1 + interactionOMOreserves.l1 + interactionMROreserves.l1 + interactioncreditreservesloans.l1 + MRO.l2 + OMO.l2 + reserves.l2 + Householdloans.l2 + SovCISS.l2 + unemployment.l2 + finalconsumption.l2 + interactionOMOreserves.l2 + interactionMROreserves.l2 + interactioncreditreservesloans.l2 + MRO.l3 + OMO.l3 + reserves.l3 + Householdloans.l3 + SovCISS.l3 + unemployment.l3 + finalconsumption.l3 + interactionOMOreserves.l3 + interactionMROreserves.l3 + interactioncreditreservesloans.l3 + const
##
## MRO.l1 OMO.l1
## -9.367919e-02 2.656792e-05
## reserves.l1 Householdloans.l1
## -4.399840e-07 -2.518485e-01
## SovCISS.l1 unemployment.l1
## 8.410062e-01 2.973028e-02
## finalconsumption.l1 interactionOMOreserves.l1
## 2.363888e-07 2.095182e-08
## interactionMROreserves.l1 interactioncreditreservesloans.l1
## -1.581941e-11 9.957423e-08
## MRO.l2 OMO.l2
## -1.585690e-02 4.525534e-05
## reserves.l2 Householdloans.l2
## 3.348078e-07 1.766766e-01
## SovCISS.l2 unemployment.l2
## -3.576603e-01 4.100073e-02
## finalconsumption.l2 interactionOMOreserves.l2
## -4.616150e-07 -1.785792e-08
## interactionMROreserves.l2 interactioncreditreservesloans.l2
## 1.178563e-12 -7.096339e-08
## MRO.l3 OMO.l3
## 4.070417e-02 1.419095e-05
## reserves.l3 Householdloans.l3
## -3.456431e-08 1.697179e-02
## SovCISS.l3 unemployment.l3
## 4.469603e-01 -1.286257e-02
## finalconsumption.l3 interactionOMOreserves.l3
## 6.277897e-07 -6.588510e-09
## interactionMROreserves.l3 interactioncreditreservesloans.l3
## -6.645637e-12 1.162951e-08
## const
## -9.116446e-01
##
##
## Estimated coefficients for equation unemployment:
## =================================================
## Call:
## unemployment = MRO.l1 + OMO.l1 + reserves.l1 + Householdloans.l1 + SovCISS.l1 + unemployment.l1 + finalconsumption.l1 + interactionOMOreserves.l1 + interactionMROreserves.l1 + interactioncreditreservesloans.l1 + MRO.l2 + OMO.l2 + reserves.l2 + Householdloans.l2 + SovCISS.l2 + unemployment.l2 + finalconsumption.l2 + interactionOMOreserves.l2 + interactionMROreserves.l2 + interactioncreditreservesloans.l2 + MRO.l3 + OMO.l3 + reserves.l3 + Householdloans.l3 + SovCISS.l3 + unemployment.l3 + finalconsumption.l3 + interactionOMOreserves.l3 + interactionMROreserves.l3 + interactioncreditreservesloans.l3 + const
##
## MRO.l1 OMO.l1
## 9.835568e-01 4.739510e-04
## reserves.l1 Householdloans.l1
## 1.191088e-06 5.410269e-01
## SovCISS.l1 unemployment.l1
## 1.378554e-01 5.383361e-01
## finalconsumption.l1 interactionOMOreserves.l1
## -8.403864e-08 5.535917e-08
## interactionMROreserves.l1 interactioncreditreservesloans.l1
## -2.452010e-10 -1.961894e-07
## MRO.l2 OMO.l2
## -8.310082e-01 -1.611666e-04
## reserves.l2 Householdloans.l2
## 3.209974e-07 -5.010887e-02
## SovCISS.l2 unemployment.l2
## 8.616446e-01 -6.104912e-01
## finalconsumption.l2 interactionOMOreserves.l2
## -2.755920e-06 9.924174e-08
## interactionMROreserves.l2 interactioncreditreservesloans.l2
## -1.755578e-10 -6.899233e-08
## MRO.l3 OMO.l3
## -1.564215e-02 1.016255e-03
## reserves.l3 Householdloans.l3
## -1.753884e-07 -7.763925e-01
## SovCISS.l3 unemployment.l3
## -2.319638e+00 -2.667143e-02
## finalconsumption.l3 interactionOMOreserves.l3
## 2.543024e-07 -3.468979e-07
## interactionMROreserves.l3 interactioncreditreservesloans.l3
## -1.351457e-10 1.628151e-07
## const
## 1.165409e+01
##
##
## Estimated coefficients for equation finalconsumption:
## =====================================================
## Call:
## finalconsumption = MRO.l1 + OMO.l1 + reserves.l1 + Householdloans.l1 + SovCISS.l1 + unemployment.l1 + finalconsumption.l1 + interactionOMOreserves.l1 + interactionMROreserves.l1 + interactioncreditreservesloans.l1 + MRO.l2 + OMO.l2 + reserves.l2 + Householdloans.l2 + SovCISS.l2 + unemployment.l2 + finalconsumption.l2 + interactionOMOreserves.l2 + interactionMROreserves.l2 + interactioncreditreservesloans.l2 + MRO.l3 + OMO.l3 + reserves.l3 + Householdloans.l3 + SovCISS.l3 + unemployment.l3 + finalconsumption.l3 + interactionOMOreserves.l3 + interactionMROreserves.l3 + interactioncreditreservesloans.l3 + const
##
## MRO.l1 OMO.l1
## 7.271596e+04 3.071775e+01
## reserves.l1 Householdloans.l1
## 5.744981e-02 7.940088e+04
## SovCISS.l1 unemployment.l1
## 2.031745e+05 -6.893902e+03
## finalconsumption.l1 interactionOMOreserves.l1
## 7.879699e-01 -6.798070e-03
## interactionMROreserves.l1 interactioncreditreservesloans.l1
## 2.483684e-05 -2.583023e-02
## MRO.l2 OMO.l2
## -2.977523e+04 -8.681595e+00
## reserves.l2 Householdloans.l2
## 5.302765e-02 6.509449e+04
## SovCISS.l2 unemployment.l2
## -1.940887e+03 -5.986889e+03
## finalconsumption.l2 interactionOMOreserves.l2
## -9.821163e-02 5.265233e-03
## interactionMROreserves.l2 interactioncreditreservesloans.l2
## 2.068057e-06 -1.510898e-02
## MRO.l3 OMO.l3
## 2.660591e+04 6.784324e+01
## reserves.l3 Householdloans.l3
## -1.391646e-01 -1.357653e+05
## SovCISS.l3 unemployment.l3
## -1.455203e+05 -2.425590e+04
## finalconsumption.l3 interactionOMOreserves.l3
## -2.120366e-01 -1.841008e-02
## interactionMROreserves.l3 interactioncreditreservesloans.l3
## -2.766081e-05 4.600827e-02
## const
## 9.441980e+05
##
##
## Estimated coefficients for equation interactionOMOreserves:
## ===========================================================
## Call:
## interactionOMOreserves = MRO.l1 + OMO.l1 + reserves.l1 + Householdloans.l1 + SovCISS.l1 + unemployment.l1 + finalconsumption.l1 + interactionOMOreserves.l1 + interactionMROreserves.l1 + interactioncreditreservesloans.l1 + MRO.l2 + OMO.l2 + reserves.l2 + Householdloans.l2 + SovCISS.l2 + unemployment.l2 + finalconsumption.l2 + interactionOMOreserves.l2 + interactionMROreserves.l2 + interactioncreditreservesloans.l2 + MRO.l3 + OMO.l3 + reserves.l3 + Householdloans.l3 + SovCISS.l3 + unemployment.l3 + finalconsumption.l3 + interactionOMOreserves.l3 + interactionMROreserves.l3 + interactioncreditreservesloans.l3 + const
##
## MRO.l1 OMO.l1
## 6.073781e+06 -1.519717e+03
## reserves.l1 Householdloans.l1
## 1.437484e+01 7.633607e+06
## SovCISS.l1 unemployment.l1
## -1.280579e+07 -1.188952e+06
## finalconsumption.l1 interactionOMOreserves.l1
## 1.042998e+00 -1.254074e+00
## interactionMROreserves.l1 interactioncreditreservesloans.l1
## -2.852743e-04 -2.615306e+00
## MRO.l2 OMO.l2
## -2.244154e+06 -1.587772e+03
## reserves.l2 Householdloans.l2
## -5.043852e+00 -2.248277e+06
## SovCISS.l2 unemployment.l2
## 1.256297e+07 7.177191e+05
## finalconsumption.l2 interactionOMOreserves.l2
## 1.512486e+00 7.153362e-01
## interactionMROreserves.l2 interactioncreditreservesloans.l2
## -1.278496e-03 1.631650e+00
## MRO.l3 OMO.l3
## 1.996893e+06 4.832851e+03
## reserves.l3 Householdloans.l3
## 5.766362e+00 -2.299971e+06
## SovCISS.l3 unemployment.l3
## -2.194306e+06 -7.113261e+05
## finalconsumption.l3 interactionOMOreserves.l3
## -8.102030e+00 -1.442352e+00
## interactionMROreserves.l3 interactioncreditreservesloans.l3
## -3.963356e-03 9.838102e-01
## const
## -8.849397e+06
##
##
## Estimated coefficients for equation interactionMROreserves:
## ===========================================================
## Call:
## interactionMROreserves = MRO.l1 + OMO.l1 + reserves.l1 + Householdloans.l1 + SovCISS.l1 + unemployment.l1 + finalconsumption.l1 + interactionOMOreserves.l1 + interactionMROreserves.l1 + interactioncreditreservesloans.l1 + MRO.l2 + OMO.l2 + reserves.l2 + Householdloans.l2 + SovCISS.l2 + unemployment.l2 + finalconsumption.l2 + interactionOMOreserves.l2 + interactionMROreserves.l2 + interactioncreditreservesloans.l2 + MRO.l3 + OMO.l3 + reserves.l3 + Householdloans.l3 + SovCISS.l3 + unemployment.l3 + finalconsumption.l3 + interactionOMOreserves.l3 + interactionMROreserves.l3 + interactioncreditreservesloans.l3 + const
##
## MRO.l1 OMO.l1
## 2.678995e+09 -9.433815e+05
## reserves.l1 Householdloans.l1
## 5.589753e+03 3.819660e+09
## SovCISS.l1 unemployment.l1
## -6.882471e+09 -1.397865e+09
## finalconsumption.l1 interactionOMOreserves.l1
## 1.228514e+03 -8.242433e+02
## interactionMROreserves.l1 interactioncreditreservesloans.l1
## 4.393681e-01 -1.251797e+03
## MRO.l2 OMO.l2
## -9.467690e+08 -1.191949e+06
## reserves.l2 Householdloans.l2
## -9.233770e+02 -1.416443e+09
## SovCISS.l2 unemployment.l2
## 3.887538e+09 1.106194e+09
## finalconsumption.l2 interactionOMOreserves.l2
## -4.648415e+03 -8.878132e+02
## interactionMROreserves.l2 interactioncreditreservesloans.l2
## -1.098615e-02 4.395458e+02
## MRO.l3 OMO.l3
## 6.554338e+08 2.359632e+06
## reserves.l3 Householdloans.l3
## 2.965512e+03 -1.074700e+09
## SovCISS.l3 unemployment.l3
## 3.342604e+09 -8.362490e+08
## finalconsumption.l3 interactionOMOreserves.l3
## -1.755419e+03 -1.295254e+03
## interactionMROreserves.l3 interactioncreditreservesloans.l3
## -1.637489e+00 4.775872e+02
## const
## 6.164527e+09
##
##
## Estimated coefficients for equation interactioncreditreservesloans:
## ===================================================================
## Call:
## interactioncreditreservesloans = MRO.l1 + OMO.l1 + reserves.l1 + Householdloans.l1 + SovCISS.l1 + unemployment.l1 + finalconsumption.l1 + interactionOMOreserves.l1 + interactionMROreserves.l1 + interactioncreditreservesloans.l1 + MRO.l2 + OMO.l2 + reserves.l2 + Householdloans.l2 + SovCISS.l2 + unemployment.l2 + finalconsumption.l2 + interactionOMOreserves.l2 + interactionMROreserves.l2 + interactioncreditreservesloans.l2 + MRO.l3 + OMO.l3 + reserves.l3 + Householdloans.l3 + SovCISS.l3 + unemployment.l3 + finalconsumption.l3 + interactionOMOreserves.l3 + interactionMROreserves.l3 + interactioncreditreservesloans.l3 + const
##
## MRO.l1 OMO.l1
## 3.410528e+06 -1.149528e+03
## reserves.l1 Householdloans.l1
## 4.418787e+00 5.069794e+06
## SovCISS.l1 unemployment.l1
## -9.854789e+06 -2.532277e+06
## finalconsumption.l1 interactionOMOreserves.l1
## 1.297284e+01 -1.226573e+00
## interactionMROreserves.l1 interactioncreditreservesloans.l1
## 1.269211e-03 -1.346980e+00
## MRO.l2 OMO.l2
## -1.660605e+06 -1.500677e+03
## reserves.l2 Householdloans.l2
## 3.066015e-01 -2.001108e+06
## SovCISS.l2 unemployment.l2
## 7.263317e+06 1.564988e+06
## finalconsumption.l2 interactionOMOreserves.l2
## -1.206300e+01 -2.049611e+00
## interactionMROreserves.l2 interactioncreditreservesloans.l2
## -5.693939e-04 6.355867e-01
## MRO.l3 OMO.l3
## 3.579092e+05 2.532786e+03
## reserves.l3 Householdloans.l3
## 6.073685e+00 -1.750689e+06
## SovCISS.l3 unemployment.l3
## 5.202567e+06 -1.346308e+06
## finalconsumption.l3 interactionOMOreserves.l3
## -3.275872e+00 -2.482156e+00
## interactionMROreserves.l3 interactioncreditreservesloans.l3
## -2.739994e-03 6.264049e-01
## const
## 9.519648e+06## Length Class Mode
## deterministic 10 -none- numeric
## A 2 -none- list
## p 1 -none- numeric
## K 1 -none- numeric
## y 265 -none- numeric
## obs 1 -none- numeric
## totobs 1 -none- numeric
## call 3 -none- call
## vecm 1 ca.jo S4
## datamat 867 -none- numeric
## resid 255 -none- numeric
## r 1 -none- numericExcept for Figure 38 the relationships of the transmission channel are easy to understand. The negative reaction of bank reserves to the MRO shock fits into the transmission channel, as the time series shows positive MRO shocks. The later reaction periods are also striking. To investigate this further, I perform IRFs based on the time series breakpoints.
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 5 4 1 5
##
## $criteria
## 1 2 3 4 5
## AIC(n) -6.595749775 -6.531486988 -6.501958889 -6.9837006110 -7.0164689068
## HQ(n) -6.507861532 -6.385006583 -6.296886322 -6.7200358828 -6.6942120168
## SC(n) -6.364098310 -6.145401212 -5.961438804 -6.2887462158 -6.1670802015
## FPE(n) 0.001366581 0.001458916 0.001506414 0.0009347164 0.0009110514
## 6 7 8 9 10
## AIC(n) -6.891377957 -6.839633413 -6.812019481 -6.882823445 -6.826807467
## HQ(n) -6.510528905 -6.400192199 -6.313986106 -6.326197908 -6.211589768
## SC(n) -5.887554942 -5.681376088 -5.499327846 -5.415697500 -5.205247212
## FPE(n) 0.001043295 0.001114802 0.001168614 0.001116872 0.001220479##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 0.27175358 0.05335604
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | 2.80 6.50 8.18 11.65
## r = 0 | 18.97 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## MRO.l8 Totalbankreservesmag.l8
## MRO.l8 1.00000 1.00000
## Totalbankreservesmag.l8 39.75563 -34.94521
##
## Weights W:
## (This is the loading matrix)
##
## MRO.l8 Totalbankreservesmag.l8
## MRO.d -0.004970236 -0.013145170
## Totalbankreservesmag.d -0.053358804 0.003408676
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 10 1 1 1
##
## $criteria
## 1 2 3 4 5 6
## AIC(n) 5.980493 6.116975 6.165569 6.285721 6.396666 6.304812
## HQ(n) 6.071117 6.268016 6.377027 6.557595 6.728956 6.697519
## SC(n) 6.226241 6.526557 6.738983 7.022968 7.297745 7.369724
## FPE(n) 395.814881 454.450444 478.863049 543.635257 613.945549 568.964733
## 7 8 9 10
## AIC(n) 6.109879 6.214941 6.250699 5.805731
## HQ(n) 6.563002 6.728480 6.824654 6.440103
## SC(n) 7.338624 7.607518 7.807108 7.525973
## FPE(n) 478.668140 547.872038 590.764711 398.586370##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 0.358477807 0.007459177
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | 0.34 6.50 8.18 11.65
## r = 0 | 20.31 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## OMO.l8 Totalbankreservesmag.l8
## OMO.l8 1.0 1.000
## Totalbankreservesmag.l8 -153021.9 -1219.474
##
## Weights W:
## (This is the loading matrix)
##
## OMO.l8 Totalbankreservesmag.l8
## OMO.d -4.589228e-03 -0.007730577
## Totalbankreservesmag.d 7.077284e-06 -0.000024056
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 10 1 1 5
##
## $criteria
## 1 2 3 4 5
## AIC(n) -6.493519502 -6.441384098 -6.29570467 -6.278633436 -6.650015797
## HQ(n) -6.406146390 -6.295762245 -6.09183408 -6.016514100 -6.329647720
## SC(n) -6.264076741 -6.058979497 -5.76033823 -5.590305154 -5.808725674
## FPE(n) 0.001513651 0.001596339 0.00185105 0.001890899 0.001313065
## 6 7 8 9 10
## AIC(n) -6.566264852 -6.584302717 -6.49227355 -6.687082757 -6.715338740
## HQ(n) -6.187648034 -6.147437158 -5.99715924 -6.133719715 -6.103726957
## SC(n) -5.572012889 -5.437088914 -5.19209790 -5.233945273 -5.109239415
## FPE(n) 0.001441846 0.001435525 0.00160296 0.001351089 0.001354164##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 0.19753210 0.08548003
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | 5.18 6.50 8.18 11.65
## r = 0 | 17.95 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## Totalbankreservesmag.l2 Householdloansmag.l2
## Totalbankreservesmag.l2 1.000000 1.000000
## Householdloansmag.l2 -0.645172 4.362919
##
## Weights W:
## (This is the loading matrix)
##
## Totalbankreservesmag.l2 Householdloansmag.l2
## Totalbankreservesmag.d -0.3942012 -0.03652812
## Householdloansmag.d 0.2106976 -0.08230752
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 6 6 1 6
##
## $criteria
## 1 2 3 4 5
## AIC(n) -5.954776813 -6.093561623 -6.074930891 -6.166059612 -6.209158783
## HQ(n) -5.865897285 -5.945429076 -5.867545325 -5.899421027 -5.883267179
## SC(n) -5.718587758 -5.699913197 -5.523823095 -5.457492446 -5.343132246
## FPE(n) 0.002594324 0.002261006 0.002310096 0.002119689 0.002046794
## 6 7 8 9 10
## AIC(n) -6.403387295 -6.352252804 -6.361810056 -6.242427367 -6.330726135
## HQ(n) -6.018242672 -5.907855162 -5.858159396 -5.679523688 -5.708569437
## SC(n) -5.379901387 -5.171307526 -5.023405408 -4.746563349 -4.677402746
## FPE(n) 0.001705662 0.001825226 0.001848747 0.002145145 0.002039322##
## ######################
## # Johansen-Procedure #
## ######################
##
## Test type: trace statistic , with linear trend
##
## Eigenvalues (lambda):
## [1] 0.28098229 0.08355418
##
## Values of teststatistic and critical values of test:
##
## test 10pct 5pct 1pct
## r <= 1 | 4.80 6.50 8.18 11.65
## r = 0 | 22.94 15.66 17.95 23.52
##
## Eigenvectors, normalised to first column:
## (These are the cointegration relations)
##
## Householdloansmag.l2 Finalconsumptionmag.l2
## Householdloansmag.l2 1.0000000 1.000000000
## Finalconsumptionmag.l2 -0.6359813 0.006870273
##
## Weights W:
## (This is the loading matrix)
##
## Householdloansmag.l2 Finalconsumptionmag.l2
## Householdloansmag.d 4.140293e-05 -0.29846780
## Finalconsumptionmag.d 8.584887e-01 0.02669315
##
## Optimal (m+1)-segment partition:
##
## Call:
## breakpoints.formula(formula = MRO ~ 1, data = dataeurozone)
##
## Breakpoints at observation number:
##
## m = 1 35
## m = 2 32 40
## m = 3 26 34 42
## m = 4 26 34 42 51
## m = 5 8 26 34 42 51
## m = 6 8 16 26 34 42 51
##
## Corresponding to breakdates:
##
## m = 1 0.593220338983051
## m = 2 0.542372881355932
## m = 3 0.440677966101695 0.576271186440678
## m = 4 0.440677966101695 0.576271186440678
## m = 5 0.135593220338983 0.440677966101695 0.576271186440678
## m = 6 0.135593220338983 0.271186440677966 0.440677966101695 0.576271186440678
##
## m = 1
## m = 2 0.677966101694915
## m = 3 0.711864406779661
## m = 4 0.711864406779661 0.864406779661017
## m = 5 0.711864406779661 0.864406779661017
## m = 6 0.711864406779661 0.864406779661017
##
## Fit:
##
## m 0 1 2 3 4 5 6
## RSS 220.460 16.095 8.747 8.011 7.361 7.361 7.361
## BIC 253.362 107.103 79.279 82.246 85.410 93.566 101.721##
## Optimal (m+1)-segment partition:
##
## Call:
## breakpoints.formula(formula = OMO ~ 1, data = dataeurozone)
##
## Breakpoints at observation number:
##
## m = 1 38
## m = 2 7 35
## m = 3 7 35 42
## m = 4 7 14 34 41
## m = 5 7 14 28 35 42
## m = 6 7 14 21 28 35 42
##
## Corresponding to breakdates:
##
## m = 1
## m = 2 0.132075471698113
## m = 3 0.132075471698113
## m = 4 0.132075471698113 0.264150943396226
## m = 5 0.132075471698113 0.264150943396226 0.528301886792453
## m = 6 0.132075471698113 0.264150943396226 0.39622641509434 0.528301886792453
##
## m = 1 0.716981132075472
## m = 2 0.660377358490566
## m = 3 0.660377358490566 0.792452830188679
## m = 4 0.641509433962264 0.773584905660377
## m = 5 0.660377358490566 0.792452830188679
## m = 6 0.660377358490566 0.792452830188679
##
## Fit:
##
## m 0 1 2 3 4 5 6
## RSS 2.704e+07 1.036e+07 4.804e+06 2.694e+06 1.825e+06 1.787e+06 1.783e+06
## BIC 8.549e+02 8.120e+02 7.792e+02 7.565e+02 7.438e+02 7.506e+02 7.584e+02##
## Optimal (m+1)-segment partition:
##
## Call:
## breakpoints.formula(formula = Totalbankreservesmag ~ 1, data = dataeurozone)
##
## Breakpoints at observation number:
##
## m = 1 33
## m = 2 33 45
## m = 3 16 33 45
## m = 4 16 33 42 51
## m = 5 9 18 33 42 51
##
## Corresponding to breakdates:
##
## m = 1 0.55
## m = 2 0.55 0.75
## m = 3 0.266666666666667 0.55 0.75
## m = 4 0.266666666666667 0.55 0.7 0.85
## m = 5 0.15 0.3 0.55 0.7 0.85
##
## Fit:
##
## m 0 1 2 3 4 5
## RSS 2.150 1.647 1.185 1.175 1.169 1.168
## BIC -21.258 -29.084 -40.619 -32.970 -25.079 -16.945##
## Optimal (m+1)-segment partition:
##
## Call:
## breakpoints.formula(formula = Householdloansmag ~ 1, data = dataeurozone)
##
## Breakpoints at observation number:
##
## m = 1 51
## m = 2 37 51
## m = 3 30 39 51
## m = 4 14 30 39 51
## m = 5 12 21 30 39 51
##
## Corresponding to breakdates:
##
## m = 1 0.85
## m = 2 0.616666666666667 0.85
## m = 3 0.5 0.65 0.85
## m = 4 0.233333333333333 0.5 0.65 0.85
## m = 5 0.2 0.35 0.5 0.65 0.85
##
## Fit:
##
## m 0 1 2 3 4 5
## RSS 2.399 2.253 1.913 1.894 1.882 1.884
## BIC -14.684 -10.265 -11.916 -4.304 3.511 11.750
Figure 40: IRFs of the liquidity channel to the different breakpoint. #
5. Comparison of the 4 transmission channels In the following, the
results of the four transmission channels are compared and further
contextualized. Household consumption is influenced, at least in
directly, by more than the four channels listed here.
The first regression of each transmission channel describes the
relationship between MRO and OMO and the first transmission variable.
The following table gives us a brief overview of the direct effect of
the MP variables on the transmission variables.
45 Channels MRO coeffi cient ragreed 0.4730512 MRO p-value OMO coeffi
cient 5.01e-10 *** OMO p-value -0.0006327 *** rhousing 0.4785135
8.50e-10 *** 0.000287 *** -0.0003643 *** equity 871929 3e-16 *** 1456
0.0342 * portfolio 0.000207 *** 1.280e+06 4.53e-13 *** 2.919e+03
Liquidity 5.16e-05 *** 9.605e+05 1.55e-12 *** 2.289e+03 Credit 4.70e-05
*** -276313.1 0.0013 ** 1321.5 Table 7: Coefficients and significance
values of all four transmission channels OLS regressions (1st part).
3.52e-08 *** First of all, it can be seen that each of the transmission
channels tested here is significant everywhere in the direct
relationship to the first transmission chan nel variable. Initially, a
positive correlation between expansionary MP and the first transmission
channel variable is assumed by Anton (2015) for each of the transmission
channels, but negative coefficients in relation to MRO are only found
for the credit channel. In their analysis of the financial markets,
Lombardi and Sushko (2023) show that the impact of monetary policy on
equity and bonds has changed in recent years. From mid-2021, the impact
on the correlation between the equity and bond markets became positive.
The negative influence of inflation on investor expectations dominated
the effect of growth expectations. Based on these results by Lombardi
and Sushko (2023), the positive coefficient of the MRO and equity could
be explained by the fact that the negative expectations of the financial
markets with regard to inflation were reduced by the increase in the
MRO. The marginal interaction variables of the transmission channels are
shown below. Despite the differen tiation, the variables are scaled
differently, which is why the marginal inter action variables are broken
down below. Interaction variables with the largest scaling were deleted
one by one.
##
## Call:
## lm(formula = dataeurozone$finalconsumption ~ dataeurozone$rhousing +
## dataeurozone$ragreed + (dataeurozone$MRO * dataeurozone$rhousing) +
## (dataeurozone$MRO * dataeurozone$ragreed) + (dataeurozone$OMO *
## dataeurozone$rhousing) + (dataeurozone$OMO * dataeurozone$ragreed) +
## dataeurozone$unemployment, data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -176013 -16772 7828 22779 91222
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2392775.04 398244.50 6.008 3.56e-07
## dataeurozone$rhousing -247851.83 316555.70 -0.783 0.438
## dataeurozone$ragreed 353653.50 312731.96 1.131 0.264
## dataeurozone$MRO 25350.90 77875.80 0.326 0.746
## dataeurozone$OMO -80.63 151.39 -0.533 0.597
## dataeurozone$unemployment -89976.76 17306.48 -5.199 5.26e-06
## dataeurozone$rhousing:dataeurozone$MRO 50474.22 61186.00 0.825 0.414
## dataeurozone$ragreed:dataeurozone$MRO -65325.55 58142.53 -1.124 0.267
## dataeurozone$rhousing:dataeurozone$OMO 149.25 134.61 1.109 0.274
## dataeurozone$ragreed:dataeurozone$OMO -203.20 130.86 -1.553 0.128
##
## (Intercept) ***
## dataeurozone$rhousing
## dataeurozone$ragreed
## dataeurozone$MRO
## dataeurozone$OMO
## dataeurozone$unemployment ***
## dataeurozone$rhousing:dataeurozone$MRO
## dataeurozone$ragreed:dataeurozone$MRO
## dataeurozone$rhousing:dataeurozone$OMO
## dataeurozone$ragreed:dataeurozone$OMO
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 47750 on 43 degrees of freedom
## (7 observations deleted due to missingness)
## Multiple R-squared: 0.9432, Adjusted R-squared: 0.9313
## F-statistic: 79.27 on 9 and 43 DF, p-value: < 2.2e-16##
## Call:
## lm(formula = dataeurozone$durablehs ~ dataeurozone$rhousing +
## dataeurozone$ragreed + (dataeurozone$MRO * dataeurozone$rhousing) +
## (dataeurozone$MRO * dataeurozone$ragreed) + (dataeurozone$OMO *
## dataeurozone$rhousing) + (dataeurozone$OMO * dataeurozone$ragreed) +
## dataeurozone$unemployment, data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.03541 -0.18229 0.03258 0.26195 0.68816
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.322e+01 1.062e+01 2.187 0.0430 *
## dataeurozone$rhousing -1.114e+01 6.610e+00 -1.685 0.1102
## dataeurozone$ragreed 1.049e+01 5.453e+00 1.924 0.0713 .
## dataeurozone$MRO -1.507e+00 1.911e+00 -0.788 0.4413
## dataeurozone$OMO 6.103e-04 4.759e-03 0.128 0.8995
## dataeurozone$unemployment -1.192e+00 5.599e-01 -2.129 0.0482 *
## dataeurozone$rhousing:dataeurozone$MRO 8.874e-01 1.221e+00 0.727 0.4774
## dataeurozone$ragreed:dataeurozone$MRO -7.751e-01 9.751e-01 -0.795 0.4376
## dataeurozone$rhousing:dataeurozone$OMO -3.007e-04 2.934e-03 -0.103 0.9196
## dataeurozone$ragreed:dataeurozone$OMO 3.201e-04 2.259e-03 0.142 0.8890
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5134 on 17 degrees of freedom
## (33 observations deleted due to missingness)
## Multiple R-squared: 0.9776, Adjusted R-squared: 0.9657
## F-statistic: 82.34 on 9 and 17 DF, p-value: 3.08e-12Figure 41: Interaction variables of the 1st part of each transmission channel (all variables). It is noticeable that the OMO interaction variables interact over the entire pe riod, but the transmission channels differ. The change in the coefficients of the interaction term variables from OMO to the financial wealth and house hold liquidity channel changes to negative in the middle to end of 2022. Com parable dynamics can be observed in relation to the credit channel. Here, the change in the coefficient of the interaction term takes place one year later in comparison and is not negative, but the change is very small in relation to before. The different significance levels of the variables seem to be loosely related to different patterns of the marginal interaction terms.
Furthermore, the interaction terms for MRO are more significant than
the in teraction terms for OMO, except for the liquidity channel, where
the interac tion term of the liquidity channel refers to CCI. In
addition, the interaction terms that have a break halfway through the
period are more significant. With MRO, the positive changes increase
abruptly; with OMO, the change becomes negative. The MRO probably had
more influence on the channels, except for the liquidity channel, than
the OMOs. To compare the impression with the VECM models. I perform a
Ganger causality test with the VECM model of the corresponding
transmission channel, using MRO and OMO as impulses. For this purpose,
the VECM models are transformed into a VAR model, but the stationarity
and co-integration of the variables are not taken into account.
Furthermore, the following table refers to VAR models that were only
esti mated with the impulse and the effect, or response, variable in
order to apply a suitable Granger causality test. For this reason, the
following results are only to be understood as a further
indicator.
rhousing MRO delayed MRO instant OMO delayed OMO instant 0.01073 *
0.005102 ** 0.5613 ragreed 0.005148 ** 0.001584 ** 0.02004 * 0.1591
equity 0.1682 0.4015 0.8551 0.00406 ** portfolio 0.684 0.1058 0.2969
0.6238 assets 0.7834 0.1201 0.2456 0.597 hsloans 0.777 0.08302 (.)
0.6786
0.01558 * reserves 0.08461 (.) 0.0001331 *** 0.093 (.) 0.8899 Table 8:
P-values of the Granger-causality test for the 1st part of the transmis
sion channels. 0.1749 48 Notes: If the p-value < 0.1 this can be
interpreted as a rejection of the H0 hypothesis of the Granger causality
test. So A Granger-causes B. Table 8 shows that the real interest rate
channel is very relevant for monetary policy. Furthermore, the liquidity
channel seems to be partly relevant for the transmission of OMO.
Furthermore, the credit channel is important for the transfer of MRO. It
is to note that the variables, especially from the wealth and liquidity
channel, are less significant when the time dimension is taken into
account as in a Granger causality test. The following table compares the
significance and the coefficients of the transmission channel of each
channel to the two consumption variables in the OLS regression frame.
The regressions are done with the same control varia bles, namely GDP,
unemployment and Gini. rhousing f.c. coefficient f.c. p-value d.c.
coefficient d.c. p-value -12006 0.893 -9.1366 ragreed 1.02e-12 *** 74603
0.401 8.8049 equity 3.124e-01 1.54e-14 *** -8.477e-07 3.54e-12 ***
portfolio 0.7820 -1.320e-01 6.37e-10 *** 1.071e-06 CCI 0.3805 -4469
0.149462 0.2024 hsloans 0.1913 -58179 0.0019 ** -1.7250 Table 9:
Coefficients and significance values of all four transmission chan nels
OLS regressions (2nd part). 0.00127 ** Notes: In contrast to Table 8,
here the exogenous variables of the regressions are on the y-axis and
endogenous variables on the x-axis. f.c. stands for final consumption
and d.c. stands for durable goods consumption. When comparing the
p-values, it is noticeable that the transmission channels are relevant
for different consumption variables. It is noticeable that the trans
mission variables of the wealth channel are more significant for final
con sumption than for durable goods consumption. It should also be
emphasized that CCI alone is not significant for consumption in the
context of the OLS. This impression of the liquidity channel in relation
to the other three channels is confirmed in Table 10. 49 rhousimg f.c.
delayed f.c. instant d.c. delayed d.c. instant 0.01455 * 0.3757 0.03425
* ragreed 0.09777 0.04642 * 0.6856 0.02348 * equity 0.0307 * 0.07363 (.)
0.3759 0.1291 portfolio 0.04448 * 0.03516 * 0.154 0.6721 CCI 0.08977 (.)
0.4903 0.7912 0.218 hsloans 0.5373 8.071e-06 *** 0.09473 (.) 0.3872
Table 10: P-values of the Granger-causality test for the 2nd part of the
trans mission channels. 0.03723 * Notes: In contrast to Table 9, here
the exogenous variables of the regressions are on the y-axis and
endogenous variables on the x-axis. The regressions are performed in the
same way as described in the transmission channel part, but without the
implementation interaction terms. The interaction terms are not relevant
here and distort significance for corresponding variables. If we now
summarize the results from tables 9 and 10, it appears that the real
interest rate channel and the credit channel are particularly
significant for the transmission of monetary policy in the euro area.
And the time delay of the transmission channel seems to play an
important role, as I have al ready shown in sections 4.1 to 4.4. The
time delays can be seen in Figure 46.
##
## Optimal (m+1)-segment partition:
##
## Call:
## breakpoints.formula(formula = Assetliabilities ~ OMO + MRO +
## GDP, data = dataeurozone)
##
## Breakpoints at observation number:
##
## m = 1 28
## m = 2 10 28
## m = 3 10 28 46
## m = 4 7 25 33 46
## m = 5 7 20 27 34 46
## m = 6 7 14 21 28 36 46
##
## Corresponding to breakdates:
##
## m = 1 0.528301886792453
## m = 2 0.188679245283019 0.528301886792453
## m = 3 0.188679245283019 0.528301886792453
## m = 4 0.132075471698113 0.471698113207547
## m = 5 0.132075471698113 0.377358490566038 0.509433962264151
## m = 6 0.132075471698113 0.264150943396226 0.39622641509434 0.528301886792453
##
## m = 1
## m = 2
## m = 3 0.867924528301887
## m = 4 0.622641509433962 0.867924528301887
## m = 5 0.641509433962264 0.867924528301887
## m = 6 0.679245283018868 0.867924528301887
##
## Fit:
##
## m 0 1 2 3 4 5 6
## RSS 5.619e+13 1.301e+13 8.055e+12 4.587e+12 2.236e+12 1.089e+12 1.498e+12
## BIC 1.638e+03 1.580e+03 1.575e+03 1.565e+03 1.546e+03 1.528e+03 1.565e+03
##
## Optimal (m+1)-segment partition:
##
## Call:
## breakpoints.formula(formula = hsequity ~ OMO + MRO + GDP, data = dataeurozone)
##
## Breakpoints at observation number:
##
## m = 1 16
## m = 2 7 26
## m = 3 7 21 28
## m = 4 7 22 33 46
## m = 5 7 20 27 34 46
## m = 6 7 14 21 28 36 46
##
## Corresponding to breakdates:
##
## m = 1 0.30188679245283
## m = 2 0.132075471698113 0.490566037735849
## m = 3 0.132075471698113 0.39622641509434 0.528301886792453
## m = 4 0.132075471698113 0.415094339622642
## m = 5 0.132075471698113 0.377358490566038 0.509433962264151
## m = 6 0.132075471698113 0.264150943396226 0.39622641509434 0.528301886792453
##
## m = 1
## m = 2
## m = 3
## m = 4 0.622641509433962 0.867924528301887
## m = 5 0.641509433962264 0.867924528301887
## m = 6 0.679245283018868 0.867924528301887
##
## Fit:
##
## m 0 1 2 3 4 5 6
## RSS 1.690e+13 4.746e+12 2.041e+12 1.342e+12 7.009e+11 3.946e+11 4.769e+11
## BIC 1.574e+03 1.527e+03 1.502e+03 1.499e+03 1.485e+03 1.474e+03 1.504e+03
##
## Optimal (m+1)-segment partition:
##
## Call:
## breakpoints.formula(formula = finalconsumption ~ hsequity + Assetliabilities +
## GDP, data = dataeurozone)
##
## Breakpoints at observation number:
##
## m = 1 25
## m = 2 18 38
## m = 3 12 20 39
## m = 4 12 20 30 39
## m = 5 12 20 30 39 48
## m = 6 9 17 25 33 41 49
##
## Corresponding to breakdates:
##
## m = 1 0.43859649122807
## m = 2 0.315789473684211
## m = 3 0.210526315789474 0.350877192982456
## m = 4 0.210526315789474 0.350877192982456 0.526315789473684
## m = 5 0.210526315789474 0.350877192982456 0.526315789473684
## m = 6 0.157894736842105 0.298245614035088 0.43859649122807 0.578947368421053
##
## m = 1
## m = 2 0.666666666666667
## m = 3 0.684210526315789
## m = 4 0.684210526315789
## m = 5 0.684210526315789 0.842105263157895
## m = 6 0.719298245614035 0.859649122807018
##
## Fit:
##
## m 0 1 2 3 4 5 6
## RSS 1.122e+11 5.188e+10 9.100e+09 1.276e+09 5.963e+08 4.164e+08 2.782e+08
## BIC 1.402e+03 1.378e+03 1.299e+03 1.207e+03 1.184e+03 1.184e+03 1.181e+03
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 6 6 6 7
##
## $criteria
## 1 2 3 4 5 6 7 8 9
## AIC(n) 6.804743 5.613916 3.223175 2.169419 -6.5220188 -Inf -Inf -Inf -Inf
## HQ(n) 7.650572 7.199846 5.549206 5.235551 -2.7157864 -Inf -Inf -Inf -Inf
## SC(n) 9.098399 9.914521 9.530729 10.483922 3.7994330 -Inf -Inf -Inf -Inf
## FPE(n) 930.232590 339.632650 52.954762 66.076988 0.2785424 NaN 0 0 0
## 10
## AIC(n) -Inf
## HQ(n) -Inf
## SC(n) -Inf
## FPE(n) 0## $Granger
##
## Granger causality H0: OMO do not Granger-cause rhousing
##
## data: VAR object var_model_rhousing
## F-Test = 0.81601, df1 = 6, df2 = 68, p-value = 0.5613
##
##
## $Instant
##
## H0: No instantaneous causality between: OMO and rhousing
##
## data: VAR object var_model_rhousing
## Chi-squared = 7.8268, df = 1, p-value = 0.005148## $Granger
##
## Granger causality H0: MRO do not Granger-cause rhousing
##
## data: VAR object var_model_rhousing2
## F-Test = 3.0428, df1 = 6, df2 = 68, p-value = 0.01073
##
##
## $Instant
##
## H0: No instantaneous causality between: MRO and rhousing
##
## data: VAR object var_model_rhousing2
## Chi-squared = 7.8428, df = 1, p-value = 0.005102## $Granger
##
## Granger causality H0: OMO do not Granger-cause ragreed
##
## data: VAR object var_model_rhousing
## F-Test = 1.6053, df1 = 6, df2 = 68, p-value = 0.1591
##
##
## $Instant
##
## H0: No instantaneous causality between: OMO and ragreed
##
## data: VAR object var_model_rhousing
## Chi-squared = 8.2568, df = 1, p-value = 0.00406## $Granger
##
## Granger causality H0: MRO do not Granger-cause ragreed
##
## data: VAR object var_model_rhousing2
## F-Test = 4.0473, df1 = 6, df2 = 68, p-value = 0.001584
##
##
## $Instant
##
## H0: No instantaneous causality between: MRO and ragreed
##
## data: VAR object var_model_rhousing2
## Chi-squared = 5.4084, df = 1, p-value = 0.02004## AIC(n) HQ(n) SC(n) FPE(n)
## 6 6 6 7## $Granger
##
## Granger causality H0: OMO do not Granger-cause hsequity
##
## data: VAR object var_model_rhousing
## F-Test = 0.4319, df1 = 6, df2 = 68, p-value = 0.8551
##
##
## $Instant
##
## H0: No instantaneous causality between: OMO and hsequity
##
## data: VAR object var_model_rhousing
## Chi-squared = 0.1657, df = 1, p-value = 0.684## $Granger
##
## Granger causality H0: MRO do not Granger-cause hsequity
##
## data: VAR object var_model_rhousing2
## F-Test = 1.5739, df1 = 6, df2 = 68, p-value = 0.1682
##
##
## $Instant
##
## H0: No instantaneous causality between: MRO and hsequity
##
## data: VAR object var_model_rhousing2
## Chi-squared = 0.70379, df = 1, p-value = 0.4015## $Granger
##
## Granger causality H0: OMO do not Granger-cause Assetliabilities
##
## data: VAR object var_model_rhousing
## F-Test = 0.73423, df1 = 6, df2 = 68, p-value = 0.6238
##
##
## $Instant
##
## H0: No instantaneous causality between: OMO and Assetliabilities
##
## data: VAR object var_model_rhousing
## Chi-squared = 0.075561, df = 1, p-value = 0.7834## $Granger
##
## Granger causality H0: MRO do not Granger-cause Assetliabilities
##
## data: VAR object var_model_rhousing2
## F-Test = 1.8317, df1 = 6, df2 = 68, p-value = 0.1058
##
##
## $Instant
##
## H0: No instantaneous causality between: MRO and Assetliabilities
##
## data: VAR object var_model_rhousing2
## Chi-squared = 1.088, df = 1, p-value = 0.2969## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 7 7 7 8
##
## $criteria
## 1 2 3 4 5
## AIC(n) 4.577159e+01 4.426627e+01 4.409700e+01 4.268494e+01 4.153193e+01
## HQ(n) 4.640596e+01 4.544439e+01 4.581886e+01 4.495056e+01 4.434129e+01
## SC(n) 4.749183e+01 4.746101e+01 4.876622e+01 4.882866e+01 4.915015e+01
## FPE(n) 7.690667e+19 1.885470e+19 2.096011e+19 9.360338e+18 1.047705e+19
## 6 7 8 9 10
## AIC(n) 3.684674e+01 -Inf -Inf -Inf -Inf
## HQ(n) 4.019985e+01 -Inf -Inf -Inf -Inf
## SC(n) 4.593944e+01 -Inf -Inf -Inf -Inf
## FPE(n) 1.851926e+18 NaN 0 0 0## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 7 7 7 8
##
## $criteria
## 1 2 3 4 5
## AIC(n) 1.789257e+01 1.648779e+01 1.607973e+01 1.513468e+01 1.264310e+01
## HQ(n) 1.852694e+01 1.766591e+01 1.780160e+01 1.740029e+01 1.545246e+01
## SC(n) 1.961281e+01 1.968252e+01 2.074896e+01 2.127840e+01 2.026132e+01
## FPE(n) 6.001507e+07 1.626956e+07 1.424466e+07 1.014762e+07 2.978381e+06
## 6 7 8 9 10
## AIC(n) 6.91609 -Inf -Inf -Inf -Inf
## HQ(n) 10.26920 -Inf -Inf -Inf -Inf
## SC(n) 16.00880 -Inf -Inf -Inf -Inf
## FPE(n) 185741.56349 NaN 0 0 0## $Granger
##
## Granger causality H0: OMO do not Granger-cause hsassets
##
## data: VAR object var_model_hsassets1
## F-Test = 0.76887, df1 = 6, df2 = 68, p-value = 0.597
##
##
## $Instant
##
## H0: No instantaneous causality between: OMO and hsassets
##
## data: VAR object var_model_hsassets1
## Chi-squared = 0.080249, df = 1, p-value = 0.777## $Granger
##
## Granger causality H0: MRO do not Granger-cause hsassets
##
## data: VAR object var_model_hsassets2
## F-Test = 1.7619, df1 = 6, df2 = 68, p-value = 0.1201
##
##
## $Instant
##
## H0: No instantaneous causality between: MRO and hsassets
##
## data: VAR object var_model_hsassets2
## Chi-squared = 1.3483, df = 1, p-value = 0.2456## $Granger
##
## Granger causality H0: OMO do not Granger-cause loanshsannualgrowth
##
## data: VAR object var_model_loans1
## F-Test = 2.7284, df1 = 7, df2 = 62, p-value = 0.01558
##
##
## $Instant
##
## H0: No instantaneous causality between: OMO and loanshsannualgrowth
##
## data: VAR object var_model_loans1
## Chi-squared = 2.974, df = 1, p-value = 0.08461## $Granger
##
## Granger causality H0: MRO do not Granger-cause loanshsannualgrowth
##
## data: VAR object var_model_loans2
## F-Test = 1.9098, df1 = 7, df2 = 62, p-value = 0.08302
##
##
## $Instant
##
## H0: No instantaneous causality between: MRO and loanshsannualgrowth
##
## data: VAR object var_model_loans2
## Chi-squared = 0.17167, df = 1, p-value = 0.6786## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 9 9 9 9
##
## $criteria
## 1 2 3 4 5
## AIC(n) 4.449617e+01 4.357749e+01 4.373439e+01 4.299390e+01 4.238465e+01
## HQ(n) 4.494929e+01 4.440821e+01 4.494272e+01 4.457983e+01 4.434819e+01
## SC(n) 4.572491e+01 4.583019e+01 4.701104e+01 4.729450e+01 4.770921e+01
## FPE(n) 2.130202e+19 8.926789e+18 1.188683e+19 7.410122e+18 6.659718e+18
## 6 7 8 9 10
## AIC(n) 3.781290e+01 3.264569e+01 -7.500696e+01 -Inf -Inf
## HQ(n) 4.015404e+01 3.536443e+01 -7.191062e+01 -Inf -Inf
## SC(n) 4.416142e+01 4.001816e+01 -6.661054e+01 -Inf -Inf
## FPE(n) 1.742719e+17 6.375827e+15 2.512977e-29 0 0## $Granger
##
## Granger causality H0: OMO do not Granger-cause reserves
##
## data: VAR object var_model_reserves1
## F-Test = 0.46682, df1 = 9, df2 = 50, p-value = 0.8899
##
##
## $Instant
##
## H0: No instantaneous causality between: OMO and reserves
##
## data: VAR object var_model_reserves1
## Chi-squared = 1.8401, df = 1, p-value = 0.1749## $Granger
##
## Granger causality H0: MRO do not Granger-cause reserves
##
## data: VAR object var_model_reserves2
## F-Test = 4.7693, df1 = 9, df2 = 50, p-value = 0.0001331
##
##
## $Instant
##
## H0: No instantaneous causality between: MRO and reserves
##
## data: VAR object var_model_reserves2
## Chi-squared = 2.8216, df = 1, p-value = 0.093##
## Call:
## lm(formula = dataeurozone$finalconsumption ~ dataeurozone$rhousing +
## dataeurozone$ragreed + dataeurozone$unemployment + dataeurozone$GDP +
## dataeurozone$Gini, data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -228213 -28908 1347 37415 118919
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1822569 9282921 0.196 0.845
## dataeurozone$rhousing -12006 88506 -0.136 0.893
## dataeurozone$ragreed 74603 88181 0.846 0.401
## dataeurozone$unemployment -167029 18616 -8.972 4.55e-12 ***
## dataeurozone$GDP 476001 438133 1.086 0.282
## dataeurozone$Gini 5614 132108 0.042 0.966
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 70870 on 51 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.8711, Adjusted R-squared: 0.8585
## F-statistic: 68.95 on 5 and 51 DF, p-value: < 2.2e-16##
## Call:
## lm(formula = dataeurozone$durablehs ~ dataeurozone$rhousing +
## dataeurozone$ragreed + dataeurozone$unemployment + dataeurozone$GDP +
## dataeurozone$Gini, data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.89427 -0.32339 -0.08518 0.41398 0.90498
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -216.7270 86.9347 -2.493 0.019650 *
## dataeurozone$rhousing -9.1366 0.6957 -13.132 1.02e-12 ***
## dataeurozone$ragreed 8.8049 0.7100 12.401 3.54e-12 ***
## dataeurozone$unemployment -1.4937 0.4409 -3.388 0.002335 **
## dataeurozone$GDP 17.0057 4.5289 3.755 0.000927 ***
## dataeurozone$Gini 2.9732 1.2140 2.449 0.021674 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.477 on 25 degrees of freedom
## (29 observations deleted due to missingness)
## Multiple R-squared: 0.9754, Adjusted R-squared: 0.9704
## F-statistic: 197.9 on 5 and 25 DF, p-value: < 2.2e-16##
## Call:
## lm(formula = dataeurozone$finalconsumption ~ dataeurozone$hsequity +
## dataeurozone$Assetliabilities + dataeurozone$GDP + dataeurozone$unemployment +
## dataeurozone$Gini, data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -99476 -20759 3460 18543 64554
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.312e+07 3.393e+06 6.815 1.07e-08 ***
## dataeurozone$hsequity 3.124e-01 2.938e-02 10.631 1.54e-14 ***
## dataeurozone$Assetliabilities -1.320e-01 1.739e-02 -7.591 6.37e-10 ***
## dataeurozone$GDP -8.300e+05 8.597e+04 -9.654 4.22e-13 ***
## dataeurozone$unemployment -2.127e+04 1.351e+04 -1.575 0.122
## dataeurozone$Gini -2.651e+05 4.913e+04 -5.396 1.78e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 32950 on 51 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.9722, Adjusted R-squared: 0.9694
## F-statistic: 356.1 on 5 and 51 DF, p-value: < 2.2e-16##
## Call:
## lm(formula = dataeurozone$durablehs ~ dataeurozone$hsequity +
## dataeurozone$Assetliabilities + dataeurozone$GDP + dataeurozone$unemployment +
## dataeurozone$Gini, data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.9498 -0.6926 -0.1846 0.4590 3.4864
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -7.557e+02 2.783e+02 -2.715 0.0118 *
## dataeurozone$hsequity -8.477e-07 3.030e-06 -0.280 0.7820
## dataeurozone$Assetliabilities 1.071e-06 1.199e-06 0.893 0.3805
## dataeurozone$GDP -6.771e+00 9.262e+00 -0.731 0.4715
## dataeurozone$unemployment -6.919e-01 1.265e+00 -0.547 0.5892
## dataeurozone$Gini 1.030e+01 4.035e+00 2.552 0.0172 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.276 on 25 degrees of freedom
## (29 observations deleted due to missingness)
## Multiple R-squared: 0.8236, Adjusted R-squared: 0.7883
## F-statistic: 23.34 on 5 and 25 DF, p-value: 1.123e-08##
## Call:
## lm(formula = dataeurozone$finalconsumption ~ dataeurozone$CCI +
## dataeurozone$GDP + dataeurozone$unemployment + dataeurozone$Gini,
## data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -284248 -20253 7623 40720 117054
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -10068210 10642130 -0.946 0.348488
## dataeurozone$CCI -4469 3054 -1.463 0.149462
## dataeurozone$GDP -767188 205325 -3.736 0.000465 ***
## dataeurozone$unemployment -208945 17229 -12.128 < 2e-16 ***
## dataeurozone$Gini 198162 146957 1.348 0.183361
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 80480 on 52 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.8306, Adjusted R-squared: 0.8175
## F-statistic: 63.73 on 4 and 52 DF, p-value: < 2.2e-16##
## Call:
## lm(formula = dataeurozone$durablehs ~ dataeurozone$CCI + dataeurozone$GDP +
## dataeurozone$unemployment + dataeurozone$Gini, data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.0724 -0.7717 -0.2513 0.4810 3.2775
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -650.3577 277.3851 -2.345 0.0270 *
## dataeurozone$CCI 0.2024 0.1509 1.342 0.1913
## dataeurozone$GDP -8.3040 6.8027 -1.221 0.2332
## dataeurozone$unemployment -0.3156 1.4416 -0.219 0.8284
## dataeurozone$Gini 9.2665 3.8973 2.378 0.0251 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.294 on 26 degrees of freedom
## (29 observations deleted due to missingness)
## Multiple R-squared: 0.8113, Adjusted R-squared: 0.7823
## F-statistic: 27.95 on 4 and 26 DF, p-value: 4.441e-09##
## Call:
## lm(formula = dataeurozone$finalconsumption ~ dataeurozone$loanshsannualgrowth +
## dataeurozone$GDP + dataeurozone$unemployment + dataeurozone$Gini,
## data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -252822 -21791 4625 41580 136104
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5798230 6192626 0.936 0.3534
## dataeurozone$loanshsannualgrowth -58179 17781 -3.272 0.0019 **
## dataeurozone$GDP 572698 440565 1.300 0.1994
## dataeurozone$unemployment -184755 17916 -10.312 3.55e-14 ***
## dataeurozone$Gini -46623 87290 -0.534 0.5955
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 74780 on 52 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.8537, Adjusted R-squared: 0.8425
## F-statistic: 75.87 on 4 and 52 DF, p-value: < 2.2e-16##
## Call:
## lm(formula = dataeurozone$durablehs ~ dataeurozone$loanshsannualgrowth +
## dataeurozone$GDP + dataeurozone$unemployment + dataeurozone$Gini,
## data = dataeurozone)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.1296 -0.6832 -0.4291 0.6054 2.9815
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -646.9518 175.7378 -3.681 0.00107 **
## dataeurozone$loanshsannualgrowth -1.7250 0.4773 -3.614 0.00127 **
## dataeurozone$GDP 23.7696 11.3695 2.091 0.04648 *
## dataeurozone$unemployment -0.8860 1.0189 -0.870 0.39246
## dataeurozone$Gini 8.6558 2.4930 3.472 0.00182 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.092 on 26 degrees of freedom
## (29 observations deleted due to missingness)
## Multiple R-squared: 0.8657, Adjusted R-squared: 0.845
## F-statistic: 41.9 on 4 and 26 DF, p-value: 5.666e-11## $Granger
##
## Granger causality H0: rhousing do not Granger-cause finalconsumption
##
## data: VAR object var_model_rhousing_fcons
## F-Test = 2.8837, df1 = 6, df2 = 68, p-value = 0.01455
##
##
## $Instant
##
## H0: No instantaneous causality between: rhousing and finalconsumption
##
## data: VAR object var_model_rhousing_fcons
## Chi-squared = 0.78464, df = 1, p-value = 0.3757## $Granger
##
## Granger causality H0: ragreed do not Granger-cause finalconsumption
##
## data: VAR object var_model_ragreed_fcons
## F-Test = 2.2746, df1 = 6, df2 = 68, p-value = 0.04642
##
##
## $Instant
##
## H0: No instantaneous causality between: ragreed and finalconsumption
##
## data: VAR object var_model_ragreed_fcons
## Chi-squared = 0.16391, df = 1, p-value = 0.6856## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 3 3 3 3
##
## $criteria
## 1 2 3 4 5 6 7 8 9 10
## AIC(n) -1.851809e+01 -1.952590e+02 -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## HQ(n) -1.824526e+01 -1.947474e+02 -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## SC(n) -1.577339e+01 -1.901127e+02 -Inf -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## FPE(n) 1.593468e-08 1.837067e-82 0 0 0 0 0 0 0 0## $Granger
##
## Granger causality H0: rhousing do not Granger-cause durablehs
##
## data: VAR object var_model_rhousing_dcons
## F-Test = 3.2332, df1 = 3, df2 = 34, p-value = 0.03425
##
##
## $Instant
##
## H0: No instantaneous causality between: rhousing and durablehs
##
## data: VAR object var_model_rhousing_dcons
## Chi-squared = 2.7416, df = 1, p-value = 0.09777## $Granger
##
## Granger causality H0: ragreed do not Granger-cause durablehs
##
## data: VAR object var_model_ragreed_dcons
## F-Test = 3.5889, df1 = 3, df2 = 34, p-value = 0.02348
##
##
## $Instant
##
## H0: No instantaneous causality between: ragreed and durablehs
##
## data: VAR object var_model_ragreed_dcons
## Chi-squared = 2.303, df = 1, p-value = 0.1291## AIC(n) HQ(n) SC(n) FPE(n)
## 6 6 6 7## $Granger
##
## Granger causality H0: hsequity do not Granger-cause finalconsumption
##
## data: VAR object var_model_equity_fcons
## F-Test = 2.4928, df1 = 6, df2 = 68, p-value = 0.0307
##
##
## $Instant
##
## H0: No instantaneous causality between: hsequity and finalconsumption
##
## data: VAR object var_model_equity_fcons
## Chi-squared = 3.2002, df = 1, p-value = 0.07363## $Granger
##
## Granger causality H0: Assetliabilities do not Granger-cause
## finalconsumption
##
## data: VAR object var_model_portfolio_fcons
## F-Test = 2.4214, df1 = 6, df2 = 68, p-value = 0.03516
##
##
## $Instant
##
## H0: No instantaneous causality between: Assetliabilities and
## finalconsumption
##
## data: VAR object var_model_portfolio_fcons
## Chi-squared = 2.0322, df = 1, p-value = 0.154## AIC(n) HQ(n) SC(n) FPE(n)
## 3 3 3 3## $Granger
##
## Granger causality H0: hsequity do not Granger-cause durablehs
##
## data: VAR object var_model_equity_dcons
## F-Test = 1.0672, df1 = 3, df2 = 34, p-value = 0.3759
##
##
## $Instant
##
## H0: No instantaneous causality between: hsequity and durablehs
##
## data: VAR object var_model_equity_dcons
## Chi-squared = 4.0383, df = 1, p-value = 0.04448## $Granger
##
## Granger causality H0: Assetliabilities do not Granger-cause durablehs
##
## data: VAR object var_model_portfolio_dcons
## F-Test = 0.51891, df1 = 3, df2 = 34, p-value = 0.6721
##
##
## $Instant
##
## H0: No instantaneous causality between: Assetliabilities and durablehs
##
## data: VAR object var_model_portfolio_dcons
## Chi-squared = 2.8785, df = 1, p-value = 0.08977## AIC(n) HQ(n) SC(n) FPE(n)
## 6 6 6 7## $Granger
##
## Granger causality H0: CCI do not Granger-cause finalconsumption
##
## data: VAR object var_model_CCI_fcons
## F-Test = 0.92958, df1 = 7, df2 = 62, p-value = 0.4903
##
##
## $Instant
##
## H0: No instantaneous causality between: CCI and finalconsumption
##
## data: VAR object var_model_CCI_fcons
## Chi-squared = 0.070127, df = 1, p-value = 0.7912## AIC(n) HQ(n) SC(n) FPE(n)
## 3 3 3 3## $Granger
##
## Granger causality H0: CCI do not Granger-cause durablehs
##
## data: VAR object var_model_CCI_dcons
## F-Test = 1.5558, df1 = 3, df2 = 34, p-value = 0.218
##
##
## $Instant
##
## H0: No instantaneous causality between: CCI and durablehs
##
## data: VAR object var_model_CCI_dcons
## Chi-squared = 0.38051, df = 1, p-value = 0.5373## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 9 9 9 9
##
## $criteria
## 1 2 3 4 5
## AIC(n) 4.449617e+01 4.357749e+01 4.373439e+01 4.299390e+01 4.238465e+01
## HQ(n) 4.494929e+01 4.440821e+01 4.494272e+01 4.457983e+01 4.434819e+01
## SC(n) 4.572491e+01 4.583019e+01 4.701104e+01 4.729450e+01 4.770921e+01
## FPE(n) 2.130202e+19 8.926789e+18 1.188683e+19 7.410122e+18 6.659718e+18
## 6 7 8 9 10
## AIC(n) 3.781290e+01 3.264569e+01 -7.500696e+01 -Inf -Inf
## HQ(n) 4.015404e+01 3.536443e+01 -7.191062e+01 -Inf -Inf
## SC(n) 4.416142e+01 4.001816e+01 -6.661054e+01 -Inf -Inf
## FPE(n) 1.742719e+17 6.375827e+15 2.512977e-29 0 0## $Granger
##
## Granger causality H0: Householdloans do not Granger-cause
## finalconsumption
##
## data: VAR object var_model_reserves1
## F-Test = 6.1919, df1 = 9, df2 = 50, p-value = 8.071e-06
##
##
## $Instant
##
## H0: No instantaneous causality between: Householdloans and
## finalconsumption
##
## data: VAR object var_model_reserves1
## Chi-squared = 2.7921, df = 1, p-value = 0.09473## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 4 4 4 3
##
## $criteria
## 1 2 3 4 5 6 7 8 9 10
## AIC(n) 9.683592 5.341256 NaN -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## HQ(n) 9.829750 5.609213 NaN -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## SC(n) 11.153968 8.036946 NaN -Inf -Inf -Inf -Inf -Inf -Inf -Inf
## FPE(n) 18810.575367 715.436623 -5.756325e-65 0 0 0 0 0 0 0## $Granger
##
## Granger causality H0: Householdloans do not Granger-cause durablehs
##
## data: VAR object var_model_reserves2
## F-Test = 1.0757, df1 = 4, df2 = 28, p-value = 0.3872
##
##
## $Instant
##
## H0: No instantaneous causality between: Householdloans and durablehs
##
## data: VAR object var_model_reserves2
## Chi-squared = 4.3397, df = 1, p-value = 0.03723Figure 46: IRFs of the four transmission channels. Figure 46 shows
that the IRFs of the wealth and liquidity channels are dis tributed over
the entire time period. The IRFs of the first and last transmission
channel are mainly located in the second half of the time period. This
is partly 50 continuous with the course of the interaction terms from
Figure 42 to 44. The process of the CCI × assets interaction term is
volatile over the entire period.
The non-significance of the third transmission channel is also confirmed
by the results of the OLS regressions and the Granger tests. The wealth
channel appears to be partially significant, the first part less so and
the second part more so. The first and last transmission channels appear
to play the largest role.
5 Country comparison
In order to test the hypotheses of Uxo et al. (2024) and D it is important to implement fiscal policies in the analysis. In order to do this, I differentiate between different countries and their fiscal policy in the following. Here I focus on Spain, as the results of the analysis by Uxo et al. related to Spain. I also implement Germany and France, as these are the largest economies in the euro area. Furthermore, this makes it possible to analyze the influence of the Gini more precisely, as it differentiates more between the three economies than the time series of the aggregated data at EU level alone. The previously used data from the ECB database and Eurostat are also available at the coun try level. The fiscal measures are calculated using the ECB and eurostat data base. From the ECB data base, I use the monthly data on average nominal yields for total government debt securities in order to take into account the respective borrowing costs and market confidence. This is to take into account interest costs in response to MPs. And I use the monthly data on the total government debt securities (totalgovdebt) and holdings of debt securities is sued by the government (holddebtgov). The totalgovdebt I use implicate if debt level of the government has an influence on the monetary policy trans mission. The variable holddebtgov I use as an indicator for the impact of mon etary policy on fiscal policy via the commercial banks’ demand for govern ment bonds. When looking at the time series, the differences and similarities between the countries become clear.
The dynamics of figure 47 can be partly explained by the dynamics of
the figure 48. Thus, the dynamics in Figure 47 can be explained by
changes in demand for cash and a change in the issuance of government
bonds. The con stant slope of the total government debt securities in
the figure 48 indicates a constant increase in government debt. The rise
in demand for government bonds can therefore be partly explained by the
increase in issuance. Overall, the dynamic differs greatly from the
government debt dynamic in some cases. This indicates the influence of
changes in demand for government bonds. This may be related to changes
in interest rates, for example. Furthermore, French and Spanish bonds
have been held more by financial institutions than German government
bonds since 2022. Factors such as risk could play a role here. However,
if the risk ratings of rating agencies such as Standard & Poor’s,
Moody’s, Fitch Ratings or the Scope Group are taken into account, it
becomes clear that risk cannot be responsible for the difference in bond
holdings.27 The differences could also be due to differing fiscal
policies and the dynamics within the time series could be partly
determined by the changes in MPs. The effects of the MP on transmission
channel variables are checked in a similar way as before, only with the
implementation fiscal variable.
tcvariable = α + β1 × MRO + β2 × OMO + β3 × fiscal + fiscal × MRO +
fiscal × OMO + ɛ (14) 27 German government bonds are rated AAA by each
of the rating agencies mentioned. The French government bonds are rated
AA- or Aa2. Spanish government bonds are rated A, A- or Baa1.
(cf. Standard & Poor’s 52 The endogenous variable is a is a
placeholder in (14) tcvariable which repre sents the respective
transmission channel variables. The variable fiscal is a placeholder for
the fiscal policy variables. In the case of Germany, a high level of
government bonds held by banks has a positive effect on the impact of
MRO on the real interest rate for home loans. The same applies to the in
teraction with OMO, higher government bond holdings in combination with
increased liquidity lead to higher real interest rates. However, the
signs of the coefficients of the interaction terms of Spain and France
differ. In the case of Spain, the influence of monetary policy on the
real interest rates of household loans is reduced by the stock of
government bonds held by commercial banks. In the case of France, the
interaction terms of the first part of the first trans mission are not
significant in each case. For the first part of the financial wealth
channel, it can be seen that the negative effect of MRO on household
wealth is amplified by the change in government bonds held. The
interaction with OMO is only significant for Germany. The asset variable
of the liquidity channel is only significantly influenced by an
interaction term for Spain. The coefficient has here a negative sign.
Overall, the impression of aggregated data that the liquidity channel is
less significant via asset variables than the other transmission
channels continues here. In the first part of the credit chan nel, the
effects of MRO and OMO are synchronized with the euro area level. For
Spain, however, MRO is only significant at a 0.05 level. With the
interac tion terms for Germany and France, the interaction of MRO and
holddebtgov are significant with a positive coefficient. In Spain, only
the direct effect of OMO is highly significant.28 For the first part of
the transmission channel, it can thus be concluded that the MRO changes
are more significant for the transmission variables of Germany and
France. Furthermore, the OMO changes play a much greater role in the
interaction variables of Spain than the MRO changes. In addition, the
significant MRO interaction terms have a pos itive coefficient for
Germany and France and a negative coefficient for Spain. 28 The results
of the regressions (14) can be found in the appendix at B.21 and B.22
for the real interest rate, B.23 for wealth,
B.24 for liquidity and B.25 for the wealth channel. 53 The full
overviews of the results the second part of the transmission channels
are in the appendix from B.26 to B.33. The effect of the transmission
variable to the consumption is analyzed as for aggregate data of euro
area with the implementation of the fiscal variables. When analyzing the
second of the transmission channels, it turns out that for Germany in
particular the real in terest rate channel and credit channel are highly
significant, 0.01 level, over final consumption. For durable goods, the
liquidity channel is significant at the 0.1 level. For Spain, the wealth
channel and liquidity channel are signifi cant. Furthermore, only the
wealth channel significantly influences final con sumption at the 0.01
level. Durable goods are significant for the liquidity and credit
channel. France is mainly significantly influenced by the credit channel
and the wealth channel via durable goods. If the transmission channels
are considered as a whole for each country, the assumed transmission
channel relationships can be confirmed. Where the coefficients indicate
a different re lationship, these are again not significant. Only in the
case of the wealth chan nel for Spain is there an initial indication of
a negative effect of expansive MP on consumption, but only for the final
consumption variable, not for durable goods consumption. Since the
wealth channel according to Anton (2015) re fers to durable goods
consumption, these results are not contradictory. How ever, the
differing significance of the transmission channels for the three
countries is striking. With the implementation of the fiscal variables,
it is no ticeable that significance and interaction term coefficients
changes. In the wealth channel with the endogenous variable final
consumption, the interac tion terms from equity and MPs become more
significant. In the case of France and Germany, the effect of the wealth
transmission channel becomes more significant. In the case of Germany,
this relates to the endogenous vari able final consumption and in the
case of France to the variable durable goods consumption. In the case of
Spain, there is a decrease in the significance of the wealth channel in
relation to durable goods consumption. holdgovdebt has hardly any impact
on the liquidity channel. Only in the case of Spain does the
significance of the variables assets, CCI and the interaction term of
assets and CCI decrease, instead of being significant at a 0.01 level,
the variables are significant at a 0.05 level. The relationships do not
change significantly with 54 55
the credit channel. Although household loans have an impact on total con sumption or the consumption of durable goods in each of the three countries, only in Spain do bank reserves have a significant influence on household loans, and this does not change with the implication of holdgovdebt. How ever, the implication of holdgovdebt changes the influence of hsloans on con sumption in the case of Spain, with durable goods consumption previously at a significance level of 0.05 with a negative coefficient, to a p-value of 0.442. In the following, I use the Granger causality test to check whether the impres sions of OLS regressions are also confirmed within VAR models and whether there is an indication of time dynamics here. MRO delayed MRO instant OMO delayed OMO instant GERrhousing 0.3343 0.4917 0.2289 0.6608 GERagreed 0.322 0.3778 0.322 0.3778 GERequity 0.4539 0.5165 0.8714 0.9392 GERassets 0.8211 0.3667 0.6669 0.4557 GERhsloans 0.002318 ** 0.721 0.02917 * 0.05302 (.) GERreserves 2.193e-05 *** 0.3399 0.5049 0.5062 FRrhousing 0.3667 0.06041 (.) 0.05425 (.) 0.08548 (.) FRagreed 0.2755 0.02121 * 0.07717 (.) 0.06929 (.) FRequity 0.5384 0.2788 0.8675 0.7621 FRassets 0.6478 0.3421 0.5145 0.5097 FRhsloans 0.00408 ** 0.8735 0.006027 ** 0.007886 ** FRreserves 5.146e-06 *** 0.9761 0.8377 0.5739 ESrhousing 0.1416 0.05224 (.) 0.411 0.2632 ESagreed 0.1279 0.04609 * 0.3909 0.1846 ESequity 0.6108 0.3463 0.6504 0.5614 ESassets 0.7554 0.3241 0.6648 0.4488 EShsloans 0.1959 0.5445 0.08596 0.8809 ESreserves 2.265e-05 *** 0.02274 * 0.3941 0.8615 Table 11: Comparison of the granger-causality test results - 1st part. Notes: The data frame used in table 11 is with final consumption, the same data frames but durable goods consumption instead of final consumption is available under A.4. The portfolio variable of the wealth channel is not in cluded here as it is not available at country level. It appears that the real interest rate channel and the liquidity channel in the first part of the MTC are not as significant with final consumption as with 56
durable goods consumption. This applies in particular to the reaction
of rhousing and ragreed to MRO. Overall, the impression is confirmed
that the wealth and credit channels are the most significant.
If we compare the consumption responses of final and durable goods con
sumption from Table 12, it is implied that durable goods consumption in
Spain was more significantly influenced by the transmission variables.
f.c. deplayed f.c. instant d.c. deplayed d.c. instant GERrhousing 0.1163
0.7893 0.1311 0.3268 GERagreed 0.1621 0.7868 0.1164 0.3261 GERequity
0.004123 ** 0.02684 * 0.1049 0.1164 GERCCI 0.04171 * 0.7365 0.01163 *
0.516 GERhsloans 0.06573 (.) 0.06432 (.) 0.0002649 *** 0.5781 FRrhousing
0.2169 0.242 0.07668 (.) 0.2186 FRagreed 0.2532 0.3806 0.01517 * 0.3975
FRequity 0.354 0.1147 0.1158 0.4092 FRCCI 0.678 0.8565 0.9274 0.7881
FRhsloans 0.5323 0.1024 0.4126 0.03525 * ESrhousing 0.5652 0.06328 (.)
0.01029 * 0.8506 ESagreed 0.6295 0.09172 (.) 0.01409 * 0.701 ESequity
0.04713 * 0.5815 0.04034 * 0.5667 ESCCI 0.113 0.2943 0.04445 * 0.5879
EShsloans 0.7028 0.6324 0.08697 (.) 0.1401 Table 12: Comparison of the
granger-causality test results - 2nd part. Table 12 does not confirm
that the real interest rate is one of the most im portant, since in the
case of Germany the credit and the liquidity channel via CCI is more
significant for consumption. In France, hardly any consumption
interactions are relevant compared to the OLS regression. A difference
can be seen here between France and Germany, which appear relatively
heterogene ous in their reactions, and Spain, where durable goods
consumption seems to be much more significantly affected by MTC.
6 Conclusion
In the course of my analysis of the four transmission channels of
monetary policy, I was able to demonstrate certain dynamics. The MRO
change from mid-2022 is of particular relevance for the period 2020 to
2023. The most important transmission channels for the period for
household consumption were the real interest rate channel, particularly
in relation to durable goods consumption, and the credit channel. The
analysis using the autoregressive models shows that the wealth channel
is not significant, as neither OMO nor MRO has a significant influence
on household equity. The liquidity channel also does not appear to be
significant. The first part of the transmission chan nel in relation to
household assets is even more closely linked to the MRO, but the
transmission channel is no longer significant at the latest when the
financial distress is implemented by the CCI. Furthermore, the
significance of the transmission channels differs between European
countries, similar to the results of Duarte and Pereira (2022).
In the future, further analysis of the transmission channels will be
required. On the one hand, it is important to investigate the dynamics
with further mod els and methodologies, as the implication of more
models and data minimizes possible bias. Furthermore, the data is not
always available in monthly format, so the temporal transmission needs
to be investigated with additional data. Furthermore, the effects of
transmission channels are sometimes only signif icant in the long term;
in the literature, the real interest rate channel or the wealth channel
are mentioned more frequently. In further analysis, the long term nature
of the transmission dynamics could be part of the analysis. Fur
thermore, an analysis, for example using a fixed effects model, is
conceivable in order to further investigate the influence of fiscal
policies on the imple mentation of monetary policies in the different
European countries. Another option for analyzing the MPs 2020-2023 is to
follow the methodology out lined in Fernández (2024) for his analysis of
the credit channel in the euro area. This approach would enable the
incorporation of a network character into the analysis. In my analysis,
a relatively strict differentiation has been made between the channels
on the basis of the four MTCs. However, it should be noted that there
are some crossover effects, for example, the liquidity chan nel and the
credit channel via household loans. 57
7 Bibliography
Albacete, N., Lindner, P., 2017. How strong is the wealth channel of monetary policy transmission? A microeconometric evaluation for Austria. Monetary Policy & the Economy. Albert, J.-F., Gómez-Fernández, N., 2024. The impact of monetary policy shocks on net worth and consumption across races in the United States. Eco nomic Systems 48, 101178. https://doi.org/10.1016/j.ecosys.2023.101178 of Alp, E., Seven, Ü., 2019. The dynamics of household final consumption: The role wealth channel. Central https://doi.org/10.1016/j.cbrev.2019.03.002 Bank Review 19, 21–32. Anton, D.R., n.d. Monetary Development and Transmission in the Eurosys tem. Cui, W., Sterk, V., 2021. Quantitative easing with heterogeneous agents. Jour nal of Monetary Economics 123, 68–90. https://doi.org/10.1016/j.jmo neco.2021.07.007 De Simone, F.N., 2024. The transmission of U.S. monetary policy to small open economies. Journal of International Money and Finance 142, 103038. https://doi.org/10.1016/j.jimonfin.2024.103038 affect Du, Z., Njangang, H., Kim, Y., 2025. Beyond the target: How do monetary policies energy poverty? https://doi.org/10.1016/j.energy.2025.135079 Energy 320, 135079. Duarte, J.B., Pereira, N., 2023. The effect of monetary policy on household consumption expenditures in Portugal: A decomposition of the transmission channel. Port Econ J 22, 149–172. https://doi.org/10.1007/s10258-022 00214-1 ECB, 2025. Data categories | ECB Data Portal [WWW Document], URL https://data.ecb.europa.eu/data/data-categories (accessed 3.12.25). 58 Eggertsson, G.B., Woodford, M., Professor, 2003. Zero Bound on Interest Rates and Optimal Monetary Policy. Brookings Papers on Economic Activity 2003, 139–233. Eurostat, 2025. Database - Eurostat [WWW Document], URL https://ec.eu ropa.eu/eurostat/web/main/data/database (accessed 3.12.25). Evgenidis, A., Salachas, E., 2019. Unconventional monetary policy and the credit channel in the euro area. Economics Letters 185, 108695. https://doi.org/10.1016/j.econlet.2019.108695 Fernández Fernández, J.A., 2024. “Banking systems in the euro zone and transmission of monetary policy.” Central Bank Review 24, 100148. https://doi.org/10.1016/j.cbrev.2024.100148 Geldvermögensbildung und Außenfinanzierung in Deutschland im ersten Quartal 2024 [WWW Document], n.d. URL https://www.bundes bank.de/de/presse/pressenotizen/geldvermoegensbildung-und-aussenfinan zierung-in-deutschland-im-ersten-quartal-2024-936128 (accessed 12.15.24). Glocker, C., Wegmüller, P., 2024. Energy price surges and inflation: Fiscal policy to the rescue? Journal of International Money and Finance 149, 103201. https://doi.org/10.1016/j.jimonfin.2024.103201 Gomes, P., Seoane, H.D., 2024. Made in Europe: Monetary–Fiscal policy mix with financial frictions. European Economic Review 165, 104727. https://doi.org/10.1016/j.euroecorev.2024.104727 Hayo, B., 2023. Does the ECB’s monetary policy affect personal finances and economic inequality? A household perspective from Germany. Economic Modelling 129, 106556. https://doi.org/10.1016/j.econmod.2023.106556 Herrenbrueck, L., 2019. Frictional asset markets and the liquidity channel of monetary policy. Journal of Economic Theory 181, 82–120. https://doi.org/10.1016/j.jet.2019.02.003 Horioka, C.Y., Niimi, Y., 2020. Was the expansion of housing credit in Japan good or bad? Japan and the World Economy 53, 100996. https://doi.org/10.1016/j.japwor.2020.100996 59 Household budget survey - Microdata - Eurostat [WWW Document], n.d. URL https://ec.europa.eu/eurostat/web/microdata/household-budget-survey (accessed 2.4.25). Hughson, H., Cava, G.L., Ryan, P., Smith, P., 2016. The Household Cash Flow Channel of Monetary Policy. Kawamoto, T., Nakazawa, T., Kishaba, Y., Matsumura, K., Nakajima, J., 2023. Estimating the macroeconomic effects of Japan’s expansionary mone tary policy under Quantitative and Qualitative Monetary Easing during 2013 2020. Economic Analysis and https://doi.org/10.1016/j.eap.2023.03.007 Policy 78, 208–224. Kim, M., 2024. Declining real interest rates: The role of energy prices in en ergy importers. Energy Economics 129, 107226. https://doi.org/10.1016/j.en eco.2023.107226 Bank of China, n.d., Law of the Peoples Republic of China on the People’s Bank of China Matusche, A., Wacks, J., 2023. Does wealth inequality affect the transmission of monetary policy? Journal of Macroeconomics 75, 103474. https://doi.org/10.1016/j.jmacro.2022.103474 Mishkin, F., 1996. The Channels of Monetary Transmission: Lessons for Monetary Policy (No. w5464). National Bureau of Economic Research, Cam bridge, MA. https://doi.org/10.3386/w5464 Mishkin, F.S., 2022. The economics of money, banking, and financial mar kets, Thirteenth edition, global edition. ed. Pearson, New York. Mishkin, F.S., 2019. The economics of money, banking and financial markets, Twelfth edition, global edition. ed, The Pearson series in economics. Pearson, Harlow, England Munich. Mundra, S., Bicchal, M., 2023. Asymmetric effects of monetary policy and financial accelerator: Evidence from India. The Journal of Economic Asym metries 27, e00296. https://doi.org/10.1016/j.jeca.2023.e00296 60 Pallotti, F., Paz-Pardo, G., Slacalek, J., Tristani, O., Violante, G.L., 2024. Who bears the costs of inflation? Euro area households and the 2021–2023 shock. Journal of Monetary Economics 148, 103671. https://doi.org/10.1016/j.jmo neco.2024.103671 Purposes and Functions [WWW Document], n.d. URL http://www.pbc.gov.cn/en/3688066/3688080/index.html (accessed 3.4.25). Journal Sapriza, H., Temesvary, J., 2024. Economic activity and the bank credit chan nel. of Banking & Finance 164, 107216. https://doi.org/10.1016/j.jbankfin.2024.107216 Statistics | Eurostat [WWW Document], n.d. URL https://ec.europa.eu/euro stat/databrowser/view/namq_10_fcs/default/table?lang=en (accessed 2.4.25). Tang, P., Yang, W., 2024. Monetary policy uncertainty and financial risk: The mediating role of corporate investment. Finance Research Letters 70, 106303. https://doi.org/10.1016/j.frl.2024.106303 The Federal Reserve’s Dual Mandate - Federal Reserve Bank of Chicago [WWW Document], n.d. URL https://www.chicagofed.org/research/dual mandate/dual-mandate (accessed 3.4.25). Uxó, J., Febrero, E., Álvarez, I., 2024. Prices, markups and wages: inflation and its distributive consequences in Spain, 2021-2023. Structural Change and Economic Dynamics https://doi.org/10.1016/j.strueco.2024.09.021 S0954349X24001504. Vale, S., 2024. House prices and credit as transmission channels from mone tary policy to inequality: Evidence from OECD countries. Economic Analysis and Policy 84, 293–307. https://doi.org/10.1016/j.eap.2024.08.027
8 Appendix A: Further explanations and overview tables.
8.1 A.1: Variables and sources
Variable Source Link source Frequency HICP Eurostat Statistics | Eurostat monthly Final con sumption Eurostat Statistics | Eurostat quarterly ihousing ECB data https://data.ecb.europa.eu/data/da tasets/MIR/MIR.M.U2.B.A2C.AM.R .A.2250.EUR.N
monthly rhousing Calculated HICP - ihousing monthly iagreed ECB data https://data.ecb.europa.eu/data/da tasets/MIR/MIR.M.U2.B.L22.F.R.A. 2250.EUR.N
monthly ragreed Calculated HICP - iagreed monthly CCI Eurostat
Statistics | Eurostat monthly
MRO ECB data https://data.ecb.europa.eu/data/da
tasets/FM/FM.B.U2.EUR.4F.KR.MR R_FR.LEV
daily OMO Statista https://www.statista.com/statis tics/254133/volume-of-ecb-open market-operations/
monthly GINI ECB data https://data.ecb.europa.eu/data/da tasets/DWA/DWA.Q.I9.S14._Z._Z.N WA._Z.GI.S.N
quarterly unemploy ment Eurostat https://ec.europa.eu/eurostat/data browser/view/ei_lmhr_m__cus tom_15480338/default/table?lang=en
monthly hsloans ECB data https://data.ecb.europa.eu/data/da tasets/BSI/BSI.M.U2.N.A.A20T.A.I. U2.2250.Z01.A
monthly hsequity ECB data https://data.ecb.europa.eu/data/da tasets/SHSS/SHSS.Q.N.U2.W0.S1M. S1.N.A.LE.F511._Z._Z.XDC._T.M. V.N._T
quarterly portfolio ECB data BPS.Q.N.I9.W1.S1M.S1.LE.N.FA.P. F52._Z.EUR._T.M.N.ALL | ECB Data Portal quarterly assets ECB data https://data.ecb.europa.eu/data/da tasets/QSA/QSA.Q.N.I9.W0.S1M.S1 .N.A.LE.F._Z._Z.XDC._T.S.V.N._T
quarterly SovCISS ECB data https://data.ecb.europa.eu/data/da tasets/CISS/CISS.M.U2.Z0Z.4F.E C.SOV_EW.IDX
monthly 63
CCI Eurostat Statistics | Eurostat monthly Re serves/To tal bank re serves ECB data https://data.ecb.europa.eu/data/da tasets/IVF/IVF.M.U2.N.40.T00.A.1.Z 5.0000.Z01.E
monthly GDP/real GDP ECB data https://data.ecb.europa.eu/data/da tasets/SPF/SPF.Q.U2.RGDP.POINT. LT.Q.AVG
quarterly Bankloans/ hsloans ECB data https://data.ecb.europa.eu/data/da tasets/BSI/BSI.M.U2.Y.U.A20T.A.I. U2.2250.Z01.A
monthly GERihous ing ECB data https://data.ecb.europa.eu/data/da tasets/MIR/MIR.M.DE.B.A2C.AM.R .A.2250.EUR.N
monthly GERHICP Eurostat https://ec.europa.eu/eurostat/data browser/view/prc_hicp_mmor/de fault/table?lang=de&cate gory=prc.prc_hicp
monthly GERrhous ing ECB data GERHICP – GERihousing
monthly GERi agreed ECB data https://data.ecb.europa.eu/data/da tasets/MIR/MIR.M.DE.B.L22.A.R.A. 2250.EUR.N
monthly GER ragreed Calculated GERHICP – GERiagreed monthly FRihous ing ECB data https://data.ecb.europa.eu/data/da tasets/MIR/MIR.M.FR.B.A2C.AM.R .A.2250.EUR.N
monthly FRHICP Eurostat https://ec.europa.eu/eurostat/data browser/view/prc_hicp_mmor/de fault/table?lang=de&cate gory=prc.prc_hicp
monthly FRrhous ing Calculated FRHICP – FRihousing
monthly FRiagreed ECB data https://data.ecb.europa.eu/data/da tasets/MIR/MIR.M.FR.B.L22.A.R.A. 2250.EUR.N
monthly FRragreed Calculated FRHICP – FRiagreed monthly ESihous ing ECB data https://data.ecb.europa.eu/data/da tasets/MIR/MIR.M.ES.B.A2C.AM.R. A.2250.EUR.N
monthly ESHICP Eurostat https://ec.europa.eu/eurostat/data browser/view/prc_hicp_mmor/de fault/table?lang=de&cate gory=prc.prc_hicp
monthly ESrhous ing Calculated ESHICP - ESihousing monthly 64
ESiagreed ECB data https://data.ecb.europa.eu/data/da tasets/MIR/MIR.M.ES.B.L22.A.R.A. 2250.EUR.N
monthly ESragreed ECB data ESHICP – ESiagreed monthly GERunem ployment Eurostat https://ec.europa.eu/eurostat/data browser/view/ei_lmhr_m__cus tom_15480338/default/table?lang=en
monthly ESunem ployment Eurostat https://ec.europa.eu/eurostat/data browser/view/UNE_RT_M/de fault/table?lang=en
monthly FRunem ployment Eurostat https://ec.europa.eu/eurostat/data browser/view/UNE_RT_M/de fault/table?lang=en
monthly GERfinal consump tion Eurostat https://data.ecb.europa.eu/data/da tasets/MNA/MNA.Q.Y.ES.W0.S1M. S1.D.P31._Z._Z._T.EUR.V.N
quarterly ESfinal consump tion Eurostat https://data.ecb.europa.eu/data/da tasets/MNA/MNA.Q.Y.ES.W0.S1M. S1.D.P31._Z._Z._T.EUR.V.N
quarterly FRfinal consump tion Eurostat https://data.ecb.europa.eu/data/da tasets/MNA/MNA.Q.Y.FR.W2.S1.S1 .B.B1GQ._Z._Z._Z.EUR.LR.N
quarterly GERhsquit y
ECB data https://data.ecb.europa.eu/data/da tasets/QSA/QSA.Q.N.DE.W0.S1M.S 1.N.A.LE.F511._Z._Z.XDC._T.S.V.N ._T
quarterly FRhsequit y ECB data https://data.ecb.europa.eu/data/da tasets/QSA/QSA.Q.N.ES.W0.S1M.S 1.N.A.LE.F511._Z._Z.XDC._T.S.V.N ._T
quarterly EShsequity ECB data https://data.ecb.europa.eu/data/da tasets/QSA/QSA.Q.N.FR.W0.S1M.S 1.N.A.LE.F511._Z._Z.XDC._T.S.V.N ._T
quarterly GERCCI Eurostat https://ec.europa.eu/eurostat/data browser/view/ei_bsco_m__cus tom_15066182/default/table
monthly FRCCI Eurostat https://ec.europa.eu/eurostat/data browser/view/ei_bsco_m__cus tom_15066182/default/table
monthly ESCCI Eurostat https://ec.europa.eu/eurostat/data browser/view/ei_bsco_m__cus tom_15066182/default/table
monthly GERhsas sets ECB data https://data.ecb.europa.eu/data/da tasets/QSA/QSA.Q.N.DE.W0.S1M.S 1.N.A.LE.F._Z._Z.XDC._T.S.V.N._T quarterly 65
FRhsassets ECB data https://data.ecb.europa.eu/data/da tasets/QSA/QSA.Q.N.FR.W0.S1M.S 1.N.A.LE.F._Z._Z.XDC._T.S.V.N._T
quarterly EShsassets ECB data https://data.ecb.europa.eu/data/da tasets/QSA/QSA.Q.N.ES.W0.S1M.S 1.N.A.LE.F._Z._Z.XDC._T.S.V.N._T
quarterly FRSo vCISS ECB data https://data.ecb.europa.eu/data/da tasets/CISS/CISS.M.FR.Z0Z.4F.EC.S OV_CI.IDX
monthly GERSo vCISS ECB data https://data.ecb.europa.eu/data/da tasets/CISS/CISS.M.DE.Z0Z.4F.EC. SOV_CI.IDX
monthly ESSo vCISS ECB data https://data.ecb.europa.eu/data/da tasets/CISS/CISS.M.ES.Z0Z.4F.EC.S OV_CI.IDX
monthly GERre serves ECB data https://data.ecb.europa.eu/data/da tasets/BSI/BSI.M.DE.N.R.LRE.X.1. A1.3000.Z01.E
monthly FRreserves ECB data https://data.ecb.europa.eu/data/da tasets/BSI/BSI.M.FR.N.R.LRE.X.1. A1.3000.Z01.E
monthly ESreserves ECB data https://data.ecb.europa.eu/data/da tasets/BSI/BSI.M.ES.N.R.LRE.X.1.A 1.3000.Z01.E
monthly GERhsloa ns ECB data https://data.ecb.europa.eu/data/da tasets/BSI/BSI.M.DE.N.A.A20T.A.I. U2.2250.Z01.A
monthly EShsloans ECD data https://data.ecb.europa.eu/data/da tasets/BSI/BSI.M.ES.N.A.A20T.A.I. U2.2250.Z01.A
monthly FRhsloans ECB data https://data.ecb.europa.eu/data/da tasets/BSI/BSI.M.FR.N.A.A20T.A.I. U2.2250.Z01.A
monthly GER holddebt gov ECB data https://data.ecb.europa.eu/data/da tasets/BSI/BSI.M.DE.N.A.A30.A.1. U2.2100.Z01.E
monthly FRholddeb tgov ECB data https://data.ecb.europa.eu/data/da tasets/BSI/BSI.M.ES.N.A.A30.A.1.U 2.2100.Z01.E
monthly ESholddeb tgov ECB data https://data.ecb.europa.eu/data/da tasets/BSI/BSI.M.FR.N.A.A30.A.1.U 2.2100.Z01.E
monthly 66
8.2 A.2 Further explanations to the variables
Variable Further explanation HICP I used HICP (harmonized indices of consumption prices) in the analy sis to measure inflation. There are of course other ways to measure in flation, for example the GDP deflator, but I wanted to imply the de mand side perspective here. GINI For the GINI variable, I mainly focus on literature that sees income ine quality as a significant factor for MTCs. Accordingly, the GINI is the income Gini coefficient. OMO OMO is one of the most important variables in this analysis; all open market operations of recent years, PELTROs, TLTROs, APP and PEPP are implied here. MRO The interest rate of the ECB’s main refinancing operations is the only time series in this analysis that is available on a daily basis. Nevertheless, in order to take the monthly format into account, I have used the rate on the last day of the month. CCI I used the CCI to measure financial distress in the context of the liquidity channel, to be precise the CCI measures households’ expectations about their future financial situation. Thus, the CCI does not directly measure the financial distress of households, although the general question is how the financial distress of households can be operationalized. unemploy ment The monthly unemployment rate I am using here is according to the in ternational labor organization (ILO). According to this, unemployed is when the person has not worked in the reference week, is available to the labor market in the two weeks and has actively looked for a job in the last four weeks or will start a job in the next three months (cf, ILO, 2022). This definition does not take hidden unemployment into account. Country spe cific varia bles The country-specific variables are synchronized with the aggregated euro area variables. They are only differentiated by adding the prefix GER for Germany, FR for France and ES for Spain.
##A.3: IRFs of the 2nd part of the real interest rate channel for gas and electricity consumption.
summary(cars)## speed dist
## Min. : 4.0 Min. : 2.00
## 1st Qu.:12.0 1st Qu.: 26.00
## Median :15.0 Median : 36.00
## Mean :15.4 Mean : 42.98
## 3rd Qu.:19.0 3rd Qu.: 56.00
## Max. :25.0 Max. :120.008.3 Including Plots
You can also embed plots, for example:
Note that the echo = FALSE parameter was added to the
code chunk to prevent printing of the R code that generated the
plot.