Capstone Project - Analysis of Sovereign Debt Restructures
Vamsy Sastry
1 January 2018
Introduction
“From the days when monarchs over-borrowed for their mercantile adventures, to Argentina’s recent failure to pay its creditors, countries have long run into trouble paying back what they have borrowed. Spain’s 16th-century king, Philip II, reigned over four of his country’s defaults. Greece and Argentina have reneged on their commitments to bondholders seven and eight times respectively over the past 200 years. And most countries have defaulted at least once in their history.” - The Economist
But what exactly happens when a country is unable to pay its dues?
When a country fails to pay its creditors on time, it is said to go into “default”, the national equivalent of going bankrupt. But sovereign defaults are quite different from business bankruptcies as it is far harder for creditors to repossess the assets of a sovereign entity than to repossess the assets of a company.
Also in a globlized world where every country is interconnected, the impact of a sovereign unable to pay its creditors can have widespread and unpredictable ramifications. The
While there is no universally accepted definition, a sovereign debt restructuring can be defined as an exchange of outstanding sovereign debt instruments, such as loans or bonds, for new debt instruments or cash through a legal process. One can generally distinguish two main elements in a debt restructuring:
Debt rescheduling
This can be defined as a lengthening of maturities of the old debt, possibly involving lower interest rates. Debt reschedulings imply debt relief, as they shift contractual payments into the future
Debt reduction
This can be defined as a reduction in the face (nominal) value of the old instruments (e.g., from US$ 100 to US$ 80).
Overview of Study
Through this paper we observe the sovereign defaults in the time span of 1950-2010. We will also try and leverage additional data sources including Macroeconomic data from world bank as independent variables.
Data Sources a) Defaults - https://sites.google.com/site/christophtrebesch/data
- Macro economic data - https://data.worldbank.org/data-catalog/world-development-indicators
The analysis is broken down into 3 parts. In the first section we create the analytical data set. Next we look at relationships between the independent variables with the dependent.Lastly we run a logistic regression model to establish a mathematical relationship between the sovereign defaults and underlying macroeconomic data.
Construction of the Analytical dataset
- Data Preparation
setwd("C:/Users/srinivas.s.n/Desktop/Capstone project/Raw Data")
defaultdata.df <- read.csv(paste("Default Consolidated.csv", sep=""))
worldbank.df <- read.csv(paste("WDIData.csv", sep=""))In The first stage both the datasets are imported into R environment. Once imported the world bank dataset is transposed.Also any entries post 2010 are removed from the analysis. Since the default data dates from 1950 to 2010
library(reshape2)## Warning: package 'reshape2' was built under R version 3.4.3md <- melt(worldbank.df, id=(c("ï..Country.Name" ,"Country.Code", "Indicator.Name", "Indicator.Code")))
drop <- c("Indicator.Code")
df = md[,!(names(md) %in% drop)]
dffinal <- dcast(df,ï..Country.Name+Country.Code+variable~Indicator.Name)
dffinal$year<-dffinal$variable
dffinal$year<-substr(dffinal$year, 2, 5)
dffinal$year <- as.numeric(dffinal$year)
dffinal1 <- subset(dffinal , dffinal$year <= 2010)
save(dffinal1,file="dffinal1.Rda")write.csv(dffinal1,‘dffinal1.csv’)
Post that the default and macroeconomic datasets are merged and reuploaded
developmentdata.df <- read.csv(paste("dffinal2.csv", sep=""))The dependent variable in our final dataset is the Default Flag. This is a binary variable indicating whether the sovereign defaulted in that particular year and underwent a debt restructuring
Data Quality
In this section variables with high number of missing values are tagged and removed. For the current exercise variables with more than 40% missing are removed from the analysis
developmentdata1.df <- developmentdata.df[, -which(colMeans(is.na(developmentdata.df)) > 0.4)]Inference Out of the initial 15729 variables only 210 finally remain.
Summary Statistics
In this section we take a look at the descriptive statistics of the remaining independent variables.
library(psych)## Warning: package 'psych' was built under R version 3.4.3description.df<- describe(developmentdata1.df)Correlation between variables and Correlation with dependent variable
In this section we take a look at the correlation of variables with each other and the correlation of variables with the dependent variable which is the default flag.
developmentdata2.df <- developmentdata1.df
developmentdata2.df[is.na(developmentdata2.df)] <- 0
correlationresult.df <- cor(developmentdata2.df[sapply(developmentdata2.df, is.numeric)])write.csv(correlationresult.df,‘correlationresult.csv’)
Inference Based on preliminary investigations it is revealed that the following are the most significant drivers of sovereign defaults. However we also observe that the magnitude of correlation is rather weak.
Age.dependency.ratio….of.working.age.population. The dependency ratio is a measure showing the number of dependents, aged zero to 14 and over the age of 65, to the total population, aged 15 to 64. It is also referred to as the “total dependency ratio.” This indicator gives insight into the amount of people of nonworking age compared to the number of those of working age. Analysis indicates with increase in the Ratio the risk of sovereign default also increases
Rural.population Analysis also indicates that the risk of sovereign defaults is correspondingly higher for nations with high rural population
Merchandise.imports.from.low..and.middle.income.economies.outside.region….of.total.merchandise.imports. This variable indicates that nations which have higher imports from lower and middle income economies tend to have higher risk of default
Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. Nations which recieve substantial aid from developed nations tend to be at a higher risk of default
Mortality.rate..neonatal..per.1.000.live.births. We also observe that higher mortality rate tends to drive the probability of sovereign defaults higher.
Other noteworthy mentions include a) Emission ratings b) Surface Area c) Arable land
In the analysis we also found two variables exhibiting high correlation with sovereign default risk but the relationship is seemingly hard to explain. these variables are Mobile.cellular.subscriptions Merchandise.exports..current.US..
Logistic Regression Equation
The top 15 variables which had high correlation with the dependent were chosen for the first iteration of the logistic regression equation. However the inferences should be taken with a pinch of salt since additional checks pertaining to multicollinearity have not been conducted. Also bucketing of data will in all likelihood improve the correlation estimates and provide better drivers, but that exercise would require better data in terms of quality and coverage.
Final Variable List
- Mobile.cellular.subscriptions
- Merchandise.exports..current.US..
- Rural.population
- Age.dependency.ratio….of.working.age.population.
- Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East…North.Africa….of.total.merchandise.imports.
- Merchandise.imports.from.low..and.middle.income.economies.outside.region….of.total.merchandise.imports.
- Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US..
- Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent.
- Mortality.rate..neonatal..per.1.000.live.births.
- Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent.
- Mortality.rate..under.5..per.1.000.live.births.
- Merchandise.exports.to.economies.in.the.Arab.World….of.total.merchandise.exports.
- Mortality.rate..infant..per.1.000.live.births.
- Agricultural.land….of.land.area.
- Primary.education..duration..years.
For additional details on the variables kindly refer to the data dictionary provided
myvars <- c("Default.Flag","Mobile.cellular.subscriptions" ,"Merchandise.exports..current.US.." ,"Rural.population" ,"Age.dependency.ratio....of.working.age.population." ,"Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports." ,"Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports." ,"Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.." ,"Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent." ,"Mortality.rate..neonatal..per.1.000.live.births." ,"Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent." ,"Mortality.rate..under.5..per.1.000.live.births." ,"Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports." ,"Mortality.rate..infant..per.1.000.live.births." ,"Agricultural.land....of.land.area." ,"Primary.education..duration..years."
)
newdata <- developmentdata2.df[myvars]Summary of the Final Variables
library(psych)
describe(newdata)## vars
## Default.Flag 1
## Mobile.cellular.subscriptions 2
## Merchandise.exports..current.US.. 3
## Rural.population 4
## Age.dependency.ratio....of.working.age.population. 5
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 6
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 7
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 8
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 9
## Mortality.rate..neonatal..per.1.000.live.births. 10
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 11
## Mortality.rate..under.5..per.1.000.live.births. 12
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 13
## Mortality.rate..infant..per.1.000.live.births. 14
## Agricultural.land....of.land.area. 15
## Primary.education..duration..years. 16
## n
## Default.Flag 13464
## Mobile.cellular.subscriptions 13464
## Merchandise.exports..current.US.. 13464
## Rural.population 13464
## Age.dependency.ratio....of.working.age.population. 13464
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 13464
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 13464
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 13464
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 13464
## Mortality.rate..neonatal..per.1.000.live.births. 13464
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 13464
## Mortality.rate..under.5..per.1.000.live.births. 13464
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 13464
## Mortality.rate..infant..per.1.000.live.births. 13464
## Agricultural.land....of.land.area. 13464
## Primary.education..duration..years. 13464
## mean
## Default.Flag 1.000000e-02
## Mobile.cellular.subscriptions 2.112556e+07
## Merchandise.exports..current.US.. 1.145373e+11
## Rural.population 1.146194e+08
## Age.dependency.ratio....of.working.age.population. 6.675000e+01
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 1.730000e+00
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 9.200000e+00
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 1.226666e+08
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 5.000479e+04
## Mortality.rate..neonatal..per.1.000.live.births. 1.576000e+01
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 9.258737e+04
## Mortality.rate..under.5..per.1.000.live.births. 5.908000e+01
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 3.230000e+00
## Mortality.rate..infant..per.1.000.live.births. 3.972000e+01
## Agricultural.land....of.land.area. 3.210000e+01
## Primary.education..duration..years. 4.310000e+00
## sd
## Default.Flag 1.200000e-01
## Mobile.cellular.subscriptions 1.718217e+08
## Merchandise.exports..current.US.. 6.410787e+11
## Rural.population 3.896393e+08
## Age.dependency.ratio....of.working.age.population. 2.831000e+01
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 3.210000e+00
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 1.038000e+01
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 5.261002e+08
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 1.750564e+05
## Mortality.rate..neonatal..per.1.000.live.births. 1.956000e+01
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 3.361540e+05
## Mortality.rate..under.5..per.1.000.live.births. 7.611000e+01
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 8.080000e+00
## Mortality.rate..infant..per.1.000.live.births. 4.652000e+01
## Agricultural.land....of.land.area. 2.310000e+01
## Primary.education..duration..years. 2.550000e+00
## median
## Default.Flag 0.00
## Mobile.cellular.subscriptions 0.00
## Merchandise.exports..current.US.. 784000000.00
## Rural.population 2871358.00
## Age.dependency.ratio....of.working.age.population. 71.50
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 0.28
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 6.74
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 600000.00
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 732.33
## Mortality.rate..neonatal..per.1.000.live.births. 7.50
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 1086.57
## Mortality.rate..under.5..per.1.000.live.births. 24.70
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 0.57
## Mortality.rate..infant..per.1.000.live.births. 20.83
## Agricultural.land....of.land.area. 33.91
## Primary.education..duration..years. 6.00
## trimmed
## Default.Flag 0.000000e+00
## Mobile.cellular.subscriptions 2.100332e+05
## Merchandise.exports..current.US.. 1.007712e+10
## Rural.population 1.789658e+07
## Age.dependency.ratio....of.working.age.population. 7.043000e+01
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 1.010000e+00
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 7.490000e+00
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 1.487556e+07
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 7.347700e+03
## Mortality.rate..neonatal..per.1.000.live.births. 1.234000e+01
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 1.180869e+04
## Mortality.rate..under.5..per.1.000.live.births. 4.438000e+01
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 1.560000e+00
## Mortality.rate..infant..per.1.000.live.births. 3.184000e+01
## Agricultural.land....of.land.area. 3.109000e+01
## Primary.education..duration..years. 4.530000e+00
## mad
## Default.Flag 0.000000e+00
## Mobile.cellular.subscriptions 0.000000e+00
## Merchandise.exports..current.US.. 1.162358e+09
## Rural.population 4.213415e+06
## Age.dependency.ratio....of.working.age.population. 2.754000e+01
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 4.200000e-01
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 1.000000e+01
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 8.895600e+05
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 1.085760e+03
## Mortality.rate..neonatal..per.1.000.live.births. 1.112000e+01
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 1.610940e+03
## Mortality.rate..under.5..per.1.000.live.births. 3.662000e+01
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 8.400000e-01
## Mortality.rate..infant..per.1.000.live.births. 3.089000e+01
## Agricultural.land....of.land.area. 2.839000e+01
## Primary.education..duration..years. 1.480000e+00
## min
## Default.Flag 0
## Mobile.cellular.subscriptions 0
## Merchandise.exports..current.US.. 0
## Rural.population 0
## Age.dependency.ratio....of.working.age.population. 0
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 0
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 0
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. -105980000
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 0
## Mortality.rate..neonatal..per.1.000.live.births. 0
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 0
## Mortality.rate..under.5..per.1.000.live.births. 0
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 0
## Mortality.rate..infant..per.1.000.live.births. 0
## Agricultural.land....of.land.area. 0
## Primary.education..duration..years. 0
## max
## Default.Flag 1.000000e+00
## Mobile.cellular.subscriptions 5.295632e+09
## Merchandise.exports..current.US.. 1.626730e+13
## Rural.population 3.352151e+09
## Age.dependency.ratio....of.working.age.population. 1.208900e+02
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 4.559000e+01
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 1.000000e+02
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 9.032350e+09
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 2.242933e+06
## Mortality.rate..neonatal..per.1.000.live.births. 1.071000e+02
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 3.464398e+06
## Mortality.rate..under.5..per.1.000.live.births. 4.424000e+02
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 9.888000e+01
## Mortality.rate..infant..per.1.000.live.births. 2.738000e+02
## Agricultural.land....of.land.area. 9.344000e+01
## Primary.education..duration..years. 9.000000e+00
## range
## Default.Flag 1.000000e+00
## Mobile.cellular.subscriptions 5.295632e+09
## Merchandise.exports..current.US.. 1.626730e+13
## Rural.population 3.352151e+09
## Age.dependency.ratio....of.working.age.population. 1.208900e+02
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 4.559000e+01
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 1.000000e+02
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 9.138330e+09
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 2.242933e+06
## Mortality.rate..neonatal..per.1.000.live.births. 1.071000e+02
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 3.464398e+06
## Mortality.rate..under.5..per.1.000.live.births. 4.424000e+02
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 9.888000e+01
## Mortality.rate..infant..per.1.000.live.births. 2.738000e+02
## Agricultural.land....of.land.area. 9.344000e+01
## Primary.education..duration..years. 9.000000e+00
## skew
## Default.Flag 8.33
## Mobile.cellular.subscriptions 15.40
## Merchandise.exports..current.US.. 11.31
## Rural.population 5.05
## Age.dependency.ratio....of.working.age.population. -0.95
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 3.94
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 1.69
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 8.34
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 5.88
## Mortality.rate..neonatal..per.1.000.live.births. 1.34
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 5.72
## Mortality.rate..under.5..per.1.000.live.births. 1.56
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 6.01
## Mortality.rate..infant..per.1.000.live.births. 1.30
## Agricultural.land....of.land.area. 0.18
## Primary.education..duration..years. -0.91
## kurtosis
## Default.Flag 67.39
## Mobile.cellular.subscriptions 306.11
## Merchandise.exports..current.US.. 170.88
## Rural.population 28.48
## Age.dependency.ratio....of.working.age.population. 0.44
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 23.54
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 4.36
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 89.74
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 42.46
## Mortality.rate..neonatal..per.1.000.live.births. 1.39
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 37.50
## Mortality.rate..under.5..per.1.000.live.births. 1.99
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 44.92
## Mortality.rate..infant..per.1.000.live.births. 1.14
## Agricultural.land....of.land.area. -0.94
## Primary.education..duration..years. -0.83
## se
## Default.Flag 0.000000e+00
## Mobile.cellular.subscriptions 1.480781e+06
## Merchandise.exports..current.US.. 5.524898e+09
## Rural.population 3.357961e+06
## Age.dependency.ratio....of.working.age.population. 2.400000e-01
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 3.000000e-02
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 9.000000e-02
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 4.533999e+06
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 1.508660e+03
## Mortality.rate..neonatal..per.1.000.live.births. 1.700000e-01
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 2.897020e+03
## Mortality.rate..under.5..per.1.000.live.births. 6.600000e-01
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 7.000000e-02
## Mortality.rate..infant..per.1.000.live.births. 4.000000e-01
## Agricultural.land....of.land.area. 2.000000e-01
## Primary.education..duration..years. 2.000000e-02Corrgram
library(corrgram)## Warning: package 'corrgram' was built under R version 3.4.3corrgram(newdata, order=TRUE, lower.panel=panel.shade,
upper.panel=panel.pie,
main="Correlations")
Logistic Regression Equation
modLog = glm(Default.Flag ~. , data=newdata, family="binomial")## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurredsummary(modLog)##
## Call:
## glm(formula = Default.Flag ~ ., family = "binomial", data = newdata)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.8982 -0.1976 -0.1164 -0.0529 4.1929
##
## Coefficients:
## Estimate
## (Intercept) -6.914e+00
## Mobile.cellular.subscriptions -8.396e-07
## Merchandise.exports..current.US.. -5.655e-12
## Rural.population -2.315e-08
## Age.dependency.ratio....of.working.age.population. 1.352e-02
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 2.814e-02
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 4.127e-04
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 7.233e-10
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 2.650e-05
## Mortality.rate..neonatal..per.1.000.live.births. 5.832e-03
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. -4.313e-06
## Mortality.rate..under.5..per.1.000.live.births. 3.141e-03
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. -1.053e-02
## Mortality.rate..infant..per.1.000.live.births. -2.235e-03
## Agricultural.land....of.land.area. 1.022e-02
## Primary.education..duration..years. 2.669e-01
## Std. Error
## (Intercept) 3.857e-01
## Mobile.cellular.subscriptions 3.297e-07
## Merchandise.exports..current.US.. 3.130e-12
## Rural.population 5.379e-09
## Age.dependency.ratio....of.working.age.population. 4.045e-03
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 2.189e-02
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 9.312e-03
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 5.860e-10
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 8.676e-06
## Mortality.rate..neonatal..per.1.000.live.births. 5.043e-03
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 4.412e-06
## Mortality.rate..under.5..per.1.000.live.births. 3.842e-03
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 9.119e-03
## Mortality.rate..infant..per.1.000.live.births. 6.890e-03
## Agricultural.land....of.land.area. 3.476e-03
## Primary.education..duration..years. 4.569e-02
## z value
## (Intercept) -17.927
## Mobile.cellular.subscriptions -2.546
## Merchandise.exports..current.US.. -1.806
## Rural.population -4.303
## Age.dependency.ratio....of.working.age.population. 3.342
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 1.286
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 0.044
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 1.234
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 3.054
## Mortality.rate..neonatal..per.1.000.live.births. 1.157
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. -0.978
## Mortality.rate..under.5..per.1.000.live.births. 0.818
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. -1.155
## Mortality.rate..infant..per.1.000.live.births. -0.324
## Agricultural.land....of.land.area. 2.939
## Primary.education..duration..years. 5.841
## Pr(>|z|)
## (Intercept) < 2e-16
## Mobile.cellular.subscriptions 0.010885
## Merchandise.exports..current.US.. 0.070841
## Rural.population 1.69e-05
## Age.dependency.ratio....of.working.age.population. 0.000831
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports. 0.198485
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports. 0.964653
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US.. 0.217084
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. 0.002258
## Mortality.rate..neonatal..per.1.000.live.births. 0.247475
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent. 0.328305
## Mortality.rate..under.5..per.1.000.live.births. 0.413622
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports. 0.248041
## Mortality.rate..infant..per.1.000.live.births. 0.745618
## Agricultural.land....of.land.area. 0.003288
## Primary.education..duration..years. 5.19e-09
##
## (Intercept) ***
## Mobile.cellular.subscriptions *
## Merchandise.exports..current.US.. .
## Rural.population ***
## Age.dependency.ratio....of.working.age.population. ***
## Merchandise.imports.from.low..and.middle.income.economies.in.Middle.East...North.Africa....of.total.merchandise.imports.
## Merchandise.imports.from.low..and.middle.income.economies.outside.region....of.total.merchandise.imports.
## Net.bilateral.aid.flows.from.DAC.donors..Germany..current.US..
## Agricultural.nitrous.oxide.emissions..thousand.metric.tons.of.CO2.equivalent. **
## Mortality.rate..neonatal..per.1.000.live.births.
## Agricultural.methane.emissions..thousand.metric.tons.of.CO2.equivalent.
## Mortality.rate..under.5..per.1.000.live.births.
## Merchandise.exports.to.economies.in.the.Arab.World....of.total.merchandise.exports.
## Mortality.rate..infant..per.1.000.live.births.
## Agricultural.land....of.land.area. **
## Primary.education..duration..years. ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1962.3 on 13463 degrees of freedom
## Residual deviance: 1723.8 on 13448 degrees of freedom
## AIC: 1755.8
##
## Number of Fisher Scoring iterations: 17Significance Level of the Variables Those variables which have atleast one star in the coefficients table are sigificant. Positive coefficient means higher the value of that variable, an increased risk of default, and vice versa.
Conclusion
Banks and Financial Institutions can use this model to create a Loan Acceptance Strategy for Sovereign lending. ALso it can be used for stress testing and macro economic predictions. With availability of better data and new modelling techniques, the accuracy of this model can easily be improved. However, logistic regression remains one of the most widely used classification technique in Credit Risk Modeling.
References
- https://www.economist.com/blogs/economist-explains/2014/11/economist-explains-20 2)“Sovereign Debt Restructurings 1950-2010: Literature Survey, Data, and Stylized Facts” (with U. Das and M. Papaioannou)