setwd("d:/1.MoHinh/2022.MH.FDI.QuangNinh/data.caulenh")
library(readxl)
dulieu <-read_excel("data.fdi.qn.edited4.6.xlsx")
head(dulieu)## # A tibble: 6 × 42
## Time FDIH FDIS IGH IGS IPH IPS GDPH GDPS LAB WORK INVEST
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 2000q1 90.7 162. 699. 1114. 299. 1084. 3405. 7565904 519. 460. 78.7
## 2 2000q2 90.7 162. 691. 1065. 299. 1040. 2310. 5132327 504. 460. 77.8
## 3 2000q3 90.7 164. 713. 1076. 305. 1029. 1667. 3657358 538. 464. 79.5
## 4 2000q4 91.6 162. 699. 1114. 316. 1029. 1123. 2490688 519. 455. 79.5
## 5 2001q1 84.8 276. 898. 1532. 465. 1026. 3749. 8194138 539 475. 139.
## 6 2001q2 84.0 276. 844. 1564. 456. 1070. 2596. 5674291 528. 479. 143.
## # … with 30 more variables: TECH <dbl>, INF <dbl>, READ <dbl>, STUDENT <dbl>,
## # DOCTOR <dbl>, BED <dbl>, GINI <dbl>, LnGDP <dbl>, LnFDI <dbl>, LnIG <dbl>,
## # LnIP <dbl>, LnTECH <dbl>, LnINVEST <dbl>, LnWORK <dbl>, LnLAB <dbl>,
## # LnSTUDENT <dbl>, LnREAD <dbl>, gGDP <dbl>, gFDI <dbl>, gIG <dbl>,
## # gIP <dbl>, gTECH <dbl>, gINVEST <dbl>, gWORK <dbl>, gLAB <dbl>,
## # gSTUDENT <dbl>, gREAD <dbl>, LnGINI <dbl>, gGINI <dbl>, gEDU <dbl>dulieu <-na.omit(dulieu)dulieu <-data.frame(dulieu)
attach(dulieu)
Y1 <- cbind(LnGDP)
Y2 <- cbind(LnWORK)
Y3 <- cbind(LnTECH)
Y4 <- cbind(LnGINI)
X <- cbind(LnFDI, LnIP, INF, LnLAB, LnINVEST, LnSTUDENT)
library(cointReg)## cointReg (v0.2.0): Parameter Estimation and Inference in a Cointegrating Regression.cointRegFM(x=X,y=Y1,kernel = "qs", bandwidth = "nw")##
## ### FM-OLS model ###
##
## Model: Y1 ~ X
##
## Parameters: Kernel = "qs" // Bandwidth = 26.08361 ("Newey-West")
##
## Coefficients:
## Estimate Std.Err t value Pr(|t|>0)
## LnFDI 0.0637681 0.0037976 16.792 < 2.2e-16 ***
## LnIP 0.3012554 0.0066219 45.493 < 2.2e-16 ***
## INF -0.1108045 0.0019765 -56.060 < 2.2e-16 ***
## LnLAB 1.8991815 0.0075308 252.190 < 2.2e-16 ***
## LnINVEST 0.1425292 0.0036613 38.928 < 2.2e-16 ***
## LnSTUDENT 0.5037344 0.0285055 17.672 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1cointRegFM(x=X,y=Y2,kernel = "pa", bandwidth = "nw")##
## ### FM-OLS model ###
##
## Model: Y2 ~ X
##
## Parameters: Kernel = "pa" // Bandwidth = 28.23357 ("Newey-West")
##
## Coefficients:
## Estimate Std.Err t value Pr(|t|>0)
## LnFDI 0.0155448 0.0024473 6.3519 1.164e-08 ***
## LnIP 0.0089292 0.0042674 2.0924 0.0395305 *
## INF 0.0058945 0.0012737 4.6277 1.387e-05 ***
## LnLAB 0.9426621 0.0048531 194.2407 < 2.2e-16 ***
## LnINVEST 0.0281253 0.0023595 11.9202 < 2.2e-16 ***
## LnSTUDENT -0.0706868 0.0183699 -3.8480 0.0002365 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1cointRegFM(x=X,y=Y3,kernel = "pa", bandwidth = "and")##
## ### FM-OLS model ###
##
## Model: Y3 ~ X
##
## Parameters: Kernel = "pa" // Bandwidth = 33.81662 ("Andrews")
##
## Coefficients:
## Estimate Std.Err t value Pr(|t|>0)
## LnFDI 0.300286 0.083603 3.5918 0.0005618 ***
## LnIP 0.355160 0.145782 2.4362 0.0170341 *
## INF 0.024261 0.043514 0.5576 0.5786848
## LnLAB -1.676965 0.165789 -10.1150 5.049e-16 ***
## LnINVEST 1.053100 0.080604 13.0652 < 2.2e-16 ***
## LnSTUDENT 0.618277 0.627548 0.9852 0.3274466
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1cointRegFM(x=X,y=Y4,kernel = "ba", bandwidth = "nw")##
## ### FM-OLS model ###
##
## Model: Y4 ~ X
##
## Parameters: Kernel = "ba" // Bandwidth = 29.61635 ("Newey-West")
##
## Coefficients:
## Estimate Std.Err t value Pr(|t|>0)
## LnFDI -0.0654230 0.0066565 -9.8284 1.844e-15 ***
## LnIP 0.0223268 0.0116073 1.9235 0.0579278 .
## INF -0.0120036 0.0034646 -3.4647 0.0008515 ***
## LnLAB -0.0872469 0.0132003 -6.6095 3.785e-09 ***
## LnINVEST 0.0119918 0.0064177 1.8685 0.0653017 .
## LnSTUDENT -0.3306350 0.0499658 -6.6172 3.658e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1library(aTSA)##
## Attaching package: 'aTSA'## The following object is masked from 'package:graphics':
##
## identifyecm(Y1,X)##
## Call:
## lm(formula = dy ~ dX + ECM - 1)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.46549 -0.09137 -0.00978 0.07980 0.44940
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## dXLnFDI 0.15582 0.08591 1.814 0.073508 .
## dXLnIP 0.71440 0.14895 4.796 7.49e-06 ***
## dXINF -0.16997 0.02462 -6.903 1.14e-09 ***
## dXLnLAB 0.88703 0.67362 1.317 0.191709
## dXLnINVEST 0.51330 0.14501 3.540 0.000675 ***
## dXLnSTUDENT -0.17735 0.59424 -0.298 0.766142
## ECM -0.67548 0.10730 6.295 1.61e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1763 on 79 degrees of freedom
## Multiple R-squared: 0.796, Adjusted R-squared: 0.7779
## F-statistic: 44.02 on 7 and 79 DF, p-value: < 2.2e-16ecm(Y2,X)##
## Call:
## lm(formula = dy ~ dX + ECM - 1)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.087140 -0.012079 0.003017 0.014741 0.068405
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## dXLnFDI 0.039660 0.012795 3.100 0.00269 **
## dXLnIP 0.020915 0.022051 0.949 0.34576
## dXINF 0.004834 0.003555 1.360 0.17776
## dXLnLAB 0.460253 0.110656 4.159 8.05e-05 ***
## dXLnINVEST -0.002650 0.021345 -0.124 0.90152
## dXLnSTUDENT 0.146213 0.088357 1.655 0.10193
## ECM -0.689323 0.109616 6.289 1.65e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.02621 on 79 degrees of freedom
## Multiple R-squared: 0.4165, Adjusted R-squared: 0.3648
## F-statistic: 8.057 on 7 and 79 DF, p-value: 2.297e-07ecm(Y3,X)##
## Call:
## lm(formula = dy ~ dX + ECM - 1)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.23535 -0.05802 0.02733 0.09651 1.68439
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## dXLnFDI 0.38367 0.14357 2.672 0.00915 **
## dXLnIP 0.28205 0.24586 1.147 0.25475
## dXINF -0.02726 0.03947 -0.691 0.49188
## dXLnLAB 1.46754 1.10653 1.326 0.18858
## dXLnINVEST 0.18685 0.24274 0.770 0.44374
## dXLnSTUDENT 1.42049 0.95361 1.490 0.14031
## ECM -0.14274 0.05946 2.401 0.01872 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2918 on 79 degrees of freedom
## Multiple R-squared: 0.2338, Adjusted R-squared: 0.1659
## F-statistic: 3.444 on 7 and 79 DF, p-value: 0.002881ecm(Y4,X)##
## Call:
## lm(formula = dy ~ dX + ECM - 1)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.158518 -0.012004 0.000608 0.025193 0.181886
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## dXLnFDI -0.086064 0.026048 -3.304 0.00143 **
## dXLnIP -0.040960 0.044675 -0.917 0.36202
## dXINF -0.011380 0.007122 -1.598 0.11405
## dXLnLAB 0.016690 0.198067 0.084 0.93306
## dXLnINVEST -0.099619 0.043581 -2.286 0.02494 *
## dXLnSTUDENT -0.179832 0.171467 -1.049 0.29748
## ECM -0.502097 0.106688 4.706 1.06e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.05229 on 79 degrees of freedom
## Multiple R-squared: 0.4144, Adjusted R-squared: 0.3625
## F-statistic: 7.987 on 7 and 79 DF, p-value: 2.617e-07coint.test(Y1,X, d=1)## Response: diff(Y1,1)
## Input: diff(X,1)
## Number of inputs: 6
## Model: y ~ X - 1
## -------------------------------
## Engle-Granger Cointegration Test
## alternative: cointegrated
##
## Type 1: no trend
## lag EG p.value
## 3.00 -6.30 0.01
## -----
## Type 2: linear trend
## lag EG p.value
## 3.00 -1.08 0.10
## -----
## Type 3: quadratic trend
## lag EG p.value
## 3.00 -2.02 0.10
## -----------
## Note: p.value = 0.01 means p.value <= 0.01
## : p.value = 0.10 means p.value >= 0.10coint.test(Y2,X, d=1)## Response: diff(Y2,1)
## Input: diff(X,1)
## Number of inputs: 6
## Model: y ~ X - 1
## -------------------------------
## Engle-Granger Cointegration Test
## alternative: cointegrated
##
## Type 1: no trend
## lag EG p.value
## 3.00 -6.24 0.01
## -----
## Type 2: linear trend
## lag EG p.value
## 3.00 -0.46 0.10
## -----
## Type 3: quadratic trend
## lag EG p.value
## 3.00 3.69 0.10
## -----------
## Note: p.value = 0.01 means p.value <= 0.01
## : p.value = 0.10 means p.value >= 0.10coint.test(Y3,X, d=1)## Response: diff(Y3,1)
## Input: diff(X,1)
## Number of inputs: 6
## Model: y ~ X - 1
## -------------------------------
## Engle-Granger Cointegration Test
## alternative: cointegrated
##
## Type 1: no trend
## lag EG p.value
## 3.0000 -5.1253 0.0362
## -----
## Type 2: linear trend
## lag EG p.value
## 3.000 0.258 0.100
## -----
## Type 3: quadratic trend
## lag EG p.value
## 3.0 1.2 0.1
## -----------
## Note: p.value = 0.01 means p.value <= 0.01
## : p.value = 0.10 means p.value >= 0.10coint.test(Y4,X, d=1)## Response: diff(Y4,1)
## Input: diff(X,1)
## Number of inputs: 6
## Model: y ~ X - 1
## -------------------------------
## Engle-Granger Cointegration Test
## alternative: cointegrated
##
## Type 1: no trend
## lag EG p.value
## 3.0000 -5.2917 0.0259
## -----
## Type 2: linear trend
## lag EG p.value
## 3.00 1.63 0.10
## -----
## Type 3: quadratic trend
## lag EG p.value
## 3.00 2.21 0.10
## -----------
## Note: p.value = 0.01 means p.value <= 0.01
## : p.value = 0.10 means p.value >= 0.10LnGDP <- cbind(LnGDP)
adf.test(LnGDP)## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 0.485 0.781
## [2,] 1 0.886 0.895
## [3,] 2 1.675 0.976
## [4,] 3 30.969 0.990
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -2.60 0.0982
## [2,] 1 -1.93 0.3552
## [3,] 2 -1.61 0.4799
## [4,] 3 -2.20 0.2516
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -10.92 0.010
## [2,] 1 -12.37 0.010
## [3,] 2 -14.57 0.010
## [4,] 3 -1.02 0.929
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01adf.test(diff(LnGDP))## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -13.492 0.010
## [2,] 1 -12.493 0.010
## [3,] 2 -42.317 0.010
## [4,] 3 -0.892 0.358
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -13.52 0.01
## [2,] 1 -12.76 0.01
## [3,] 2 -161.87 0.01
## [4,] 3 -4.19 0.01
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -13.4 0.01
## [2,] 1 -12.7 0.01
## [3,] 2 -165.3 0.01
## [4,] 3 -4.2 0.01
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01LnWORK <- cbind(LnWORK)
adf.test(LnWORK)## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 1.12 0.927
## [2,] 1 1.57 0.970
## [3,] 2 2.10 0.990
## [4,] 3 2.14 0.990
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -1.89 0.370
## [2,] 1 -1.93 0.354
## [3,] 2 -2.39 0.179
## [4,] 3 -2.36 0.192
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -1.757 0.672
## [2,] 1 -0.743 0.963
## [3,] 2 0.130 0.990
## [4,] 3 0.779 0.990
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01adf.test(diff(LnWORK))## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -12.83 0.01
## [2,] 1 -9.37 0.01
## [3,] 2 -6.83 0.01
## [4,] 3 -5.09 0.01
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -13.05 0.01
## [2,] 1 -9.82 0.01
## [3,] 2 -7.33 0.01
## [4,] 3 -5.59 0.01
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -13.34 0.01
## [2,] 1 -10.53 0.01
## [3,] 2 -8.21 0.01
## [4,] 3 -6.59 0.01
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01LnTECH <- cbind(LnTECH)
adf.test(LnTECH)## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -0.0918 0.616
## [2,] 1 -0.0833 0.618
## [3,] 2 -0.0960 0.615
## [4,] 3 0.0513 0.657
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -1.79 0.411
## [2,] 1 -1.85 0.386
## [3,] 2 -1.92 0.359
## [4,] 3 -1.43 0.543
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -2.46 0.382
## [2,] 1 -2.49 0.368
## [3,] 2 -2.62 0.319
## [4,] 3 -2.43 0.396
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01adf.test(diff(LnTECH))## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -9.26 0.01
## [2,] 1 -6.38 0.01
## [3,] 2 -5.35 0.01
## [4,] 3 -3.72 0.01
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -9.43 0.01
## [2,] 1 -6.58 0.01
## [3,] 2 -5.50 0.01
## [4,] 3 -3.82 0.01
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -9.51 0.0100
## [2,] 1 -6.70 0.0100
## [3,] 2 -5.53 0.0100
## [4,] 3 -3.88 0.0194
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01LnGINI <- cbind(LnGINI)
adf.test(LnGINI)## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 0.0781 0.664
## [2,] 1 0.5991 0.813
## [3,] 2 0.5787 0.807
## [4,] 3 0.4129 0.760
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -6.02 0.01
## [2,] 1 -4.41 0.01
## [3,] 2 -5.07 0.01
## [4,] 3 -3.54 0.01
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -5.96 0.0100
## [2,] 1 -4.39 0.0100
## [3,] 2 -5.19 0.0100
## [4,] 3 -3.71 0.0288
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01adf.test(diff(LnGINI))## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -19.00 0.01
## [2,] 1 -7.99 0.01
## [3,] 2 -7.28 0.01
## [4,] 3 -6.04 0.01
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -18.98 0.01
## [2,] 1 -8.01 0.01
## [3,] 2 -7.26 0.01
## [4,] 3 -5.99 0.01
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -19.06 0.01
## [2,] 1 -8.12 0.01
## [3,] 2 -7.25 0.01
## [4,] 3 -5.89 0.01
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01LnFDI <- cbind(LnFDI)
adf.test(LnFDI)## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 0.577 0.807
## [2,] 1 0.546 0.798
## [3,] 2 0.542 0.797
## [4,] 3 0.390 0.753
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -2.87 0.0553
## [2,] 1 -3.02 0.0396
## [3,] 2 -3.22 0.0236
## [4,] 3 -2.62 0.0952
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -2.51 0.360
## [2,] 1 -2.68 0.293
## [3,] 2 -2.86 0.219
## [4,] 3 -2.39 0.411
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01adf.test(diff(LnFDI))## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -8.90 0.01
## [2,] 1 -6.27 0.01
## [3,] 2 -5.33 0.01
## [4,] 3 -3.67 0.01
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -8.92 0.01
## [2,] 1 -6.31 0.01
## [3,] 2 -5.33 0.01
## [4,] 3 -3.67 0.01
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -9.06 0.0100
## [2,] 1 -6.50 0.0100
## [3,] 2 -5.42 0.0100
## [4,] 3 -3.76 0.0244
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01LnIP <- cbind(LnIP)
adf.test(LnIP)## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 1.73 0.978
## [2,] 1 1.86 0.983
## [3,] 2 1.77 0.980
## [4,] 3 1.77 0.979
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -1.06 0.676
## [2,] 1 -1.05 0.677
## [3,] 2 -1.12 0.653
## [4,] 3 -1.16 0.639
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -1.90 0.614
## [2,] 1 -1.77 0.667
## [3,] 2 -1.83 0.643
## [4,] 3 -1.82 0.646
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01adf.test(diff(LnIP))## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -9.60 0.01
## [2,] 1 -6.17 0.01
## [3,] 2 -4.94 0.01
## [4,] 3 -3.14 0.01
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -9.95 0.010
## [2,] 1 -6.53 0.010
## [3,] 2 -5.36 0.010
## [4,] 3 -3.46 0.013
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -9.90 0.0100
## [2,] 1 -6.50 0.0100
## [3,] 2 -5.33 0.0100
## [4,] 3 -3.44 0.0534
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01INF <- cbind(INF)
adf.test(INF)## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -2.268 0.0240
## [2,] 1 -1.720 0.0839
## [3,] 2 -1.370 0.1863
## [4,] 3 -0.823 0.3828
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -3.97 0.0100
## [2,] 1 -3.07 0.0357
## [3,] 2 -2.38 0.1829
## [4,] 3 -1.35 0.5729
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -4.64 0.0100
## [2,] 1 -3.75 0.0248
## [3,] 2 -3.05 0.1422
## [4,] 3 -2.22 0.4778
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01adf.test(diff(INF))## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -12.30 0.01
## [2,] 1 -9.41 0.01
## [3,] 2 -10.85 0.01
## [4,] 3 -4.40 0.01
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -12.23 0.01
## [2,] 1 -9.35 0.01
## [3,] 2 -10.78 0.01
## [4,] 3 -4.37 0.01
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -12.18 0.01
## [2,] 1 -9.32 0.01
## [3,] 2 -10.84 0.01
## [4,] 3 -4.40 0.01
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01LnLAB <- cbind(LnLAB)
adf.test(LnLAB)## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 0.922 0.902
## [2,] 1 1.024 0.915
## [3,] 2 1.554 0.968
## [4,] 3 1.643 0.975
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -2.23 0.242
## [2,] 1 -1.78 0.414
## [3,] 2 -2.12 0.282
## [4,] 3 -2.00 0.330
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -2.204 0.485
## [2,] 1 -0.988 0.935
## [3,] 2 -0.141 0.990
## [4,] 3 0.789 0.990
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01adf.test(diff(LnLAB))## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -12.31 0.01
## [2,] 1 -9.55 0.01
## [3,] 2 -7.48 0.01
## [4,] 3 -5.33 0.01
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -12.36 0.01
## [2,] 1 -9.77 0.01
## [3,] 2 -7.74 0.01
## [4,] 3 -5.63 0.01
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -12.54 0.01
## [2,] 1 -10.25 0.01
## [3,] 2 -8.37 0.01
## [4,] 3 -6.43 0.01
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01LnINVEST <- cbind(LnINVEST)
adf.test(LnINVEST)## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 2.57 0.99
## [2,] 1 2.84 0.99
## [3,] 2 2.97 0.99
## [4,] 3 2.96 0.99
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -1.54 0.507
## [2,] 1 -1.68 0.452
## [3,] 2 -1.88 0.375
## [4,] 3 -1.25 0.609
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -2.87 0.215
## [2,] 1 -2.58 0.335
## [3,] 2 -2.45 0.384
## [4,] 3 -1.68 0.702
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01adf.test(diff(LnINVEST))## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -9.68 0.01
## [2,] 1 -6.55 0.01
## [3,] 2 -5.53 0.01
## [4,] 3 -3.82 0.01
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -10.70 0.01
## [2,] 1 -7.76 0.01
## [3,] 2 -6.70 0.01
## [4,] 3 -4.73 0.01
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -10.80 0.01
## [2,] 1 -7.94 0.01
## [3,] 2 -6.75 0.01
## [4,] 3 -4.75 0.01
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01LnSTUDENT <- cbind(LnSTUDENT)
adf.test(LnSTUDENT)## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 0.713 0.846
## [2,] 1 1.120 0.928
## [3,] 2 1.265 0.946
## [4,] 3 1.243 0.943
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -0.762 0.779
## [2,] 1 0.349 0.978
## [3,] 2 0.795 0.990
## [4,] 3 0.951 0.990
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -1.4702 0.792
## [2,] 1 -0.4735 0.981
## [3,] 2 -0.0923 0.990
## [4,] 3 0.1264 0.990
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01adf.test(diff(LnSTUDENT))## Augmented Dickey-Fuller Test
## alternative: stationary
##
## Type 1: no drift no trend
## lag ADF p.value
## [1,] 0 -13.29 0.01
## [2,] 1 -8.51 0.01
## [3,] 2 -6.22 0.01
## [4,] 3 -5.19 0.01
## Type 2: with drift no trend
## lag ADF p.value
## [1,] 0 -13.35 0.01
## [2,] 1 -8.62 0.01
## [3,] 2 -6.34 0.01
## [4,] 3 -5.34 0.01
## Type 3: with drift and trend
## lag ADF p.value
## [1,] 0 -13.72 0.01
## [2,] 1 -9.16 0.01
## [3,] 2 -7.09 0.01
## [4,] 3 -6.31 0.01
## ----
## Note: in fact, p.value = 0.01 means p.value <= 0.01