Tạo dữ liệu

library(readxl)
setwd("d:/DATA2020/PaerScopus/Von_Quy_GDP")
dulieu <-read_excel("data.cutoff.2000.xlsx")
dulieu <-data.frame(dulieu)
head(dulieu)
##   YEAR QUY   Time    GDP    LnGDP      TI     LnTI      IG     LnIG     IP
## 1 2000  Q1 2000q1 224744 12.32272 48119.6 10.78144 12415.2 9.426677 4636.6
## 2 2000  Q2 2000q2 289321 12.57529 61946.0 11.03402 15982.6 9.679256 5968.8
## 3 2000  Q3 2000q3 270389 12.50762 57892.5 10.96634 14936.8 9.611583 5578.3
## 4 2000  Q4 2000q4 330885 12.70953 70845.2 11.16825 18278.6 9.813486 6826.3
## 5 2001  Q1 2001q1 241948 12.39648 54482.6 10.90564 14573.9 9.586988 5197.9
## 6 2001  Q2 2001q2 305601 12.63004 68816.1 11.13919 18408.1 9.820546 6565.5
##       LnIP      IF     LnIF  CPI  OILP    LCU      LAB   LnOILP    LnLCU
## 1 8.441737 31067.8 10.34393  0.9 26.85 73.529 38361.71 3.290266 4.297680
## 2 8.694301 39994.6 10.59650 -0.9 29.00 69.959 38563.64 3.367296 4.247909
## 3 8.626639 37377.4 10.52882 -1.6 33.00 67.351 38924.97 3.496508 4.209918
## 4 8.828538 45740.3 10.73074 -0.5 33.76 35.419 38545.38 3.519277 3.567248
## 5 8.556010 34710.8 10.45481  0.0 27.41 34.478 39421.17 3.310908 3.540321
## 6 8.789584 43842.5 10.68836 -0.7 28.48 34.753 39629.60 3.349202 3.548266
##      LnLAB
## 1 10.55482
## 2 10.56007
## 3 10.56939
## 4 10.55959
## 5 10.58206
## 6 10.58733

Unit Root Test

library(fUnitRoots)
## Loading required package: timeDate
## Loading required package: timeSeries
## Loading required package: fBasics
library(dplyr)
## 
## Attaching package: 'dplyr'
## The following objects are masked from 'package:timeSeries':
## 
##     filter, lag
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
adfTest(dulieu$LnGDP,4)
## Warning in adfTest(dulieu$LnGDP, 4): p-value greater than printed p-value
## 
## Title:
##  Augmented Dickey-Fuller Test
## 
## Test Results:
##   PARAMETER:
##     Lag Order: 4
##   STATISTIC:
##     Dickey-Fuller: 3.402
##   P VALUE:
##     0.99 
## 
## Description:
##  Fri Jul 10 17:30:18 2020 by user: Admin
unitrootTest(dulieu$LnGDP,4)
## 
## Title:
##  Augmented Dickey-Fuller Test
## 
## Test Results:
##   PARAMETER:
##     Lag Order: 4
##   STATISTIC:
##     DF: 3.402
##   P VALUE:
##     t: 0.9998 
##     n: 0.9993 
## 
## Description:
##  Fri Jul 10 17:30:18 2020 by user: Admin

unit Root test gói khác

library(tseries)
## Registered S3 method overwritten by 'quantmod':
##   method            from
##   as.zoo.data.frame zoo
adf.test(dulieu$LnGDP)
## 
##  Augmented Dickey-Fuller Test
## 
## data:  dulieu$LnGDP
## Dickey-Fuller = -1.7261, Lag order = 4, p-value = 0.6878
## alternative hypothesis: stationary
kpss.test(dulieu$LnGDP)
## Warning in kpss.test(dulieu$LnGDP): p-value smaller than printed p-value
## 
##  KPSS Test for Level Stationarity
## 
## data:  dulieu$LnGDP
## KPSS Level = 2.1104, Truncation lag parameter = 3, p-value = 0.01
pp.test(dulieu$LnGDP)
## Warning in pp.test(dulieu$LnGDP): p-value smaller than printed p-value
## 
##  Phillips-Perron Unit Root Test
## 
## data:  dulieu$LnGDP
## Dickey-Fuller Z(alpha) = -86.91, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
adf.test(diff(dulieu$LnGDP))
## 
##  Augmented Dickey-Fuller Test
## 
## data:  diff(dulieu$LnGDP)
## Dickey-Fuller = -2.651, Lag order = 4, p-value = 0.3094
## alternative hypothesis: stationary
kpss.test(diff(dulieu$LnGDP))
## Warning in kpss.test(diff(dulieu$LnGDP)): p-value greater than printed p-value
## 
##  KPSS Test for Level Stationarity
## 
## data:  diff(dulieu$LnGDP)
## KPSS Level = 0.27903, Truncation lag parameter = 3, p-value = 0.1
pp.test(diff(dulieu$LnGDP))
## Warning in pp.test(diff(dulieu$LnGDP)): p-value smaller than printed p-value
## 
##  Phillips-Perron Unit Root Test
## 
## data:  diff(dulieu$LnGDP)
## Dickey-Fuller Z(alpha) = -105.9, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
pp.test(dulieu[,1])
## Warning in pp.test(dulieu[, 1]): p-value smaller than printed p-value
## 
##  Phillips-Perron Unit Root Test
## 
## data:  dulieu[, 1]
## Dickey-Fuller Z(alpha) = -56.216, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
head(dulieu[,1])
## [1] 2000 2000 2000 2000 2001 2001

Unit Root Test for data

dim(dulieu)
## [1] 80 20
chon <-c(5,9,11,13,14, 18,19,20)
for ( i in chon){
 print(i)
 print( pp.test(dulieu[,i]))
 print( pp.test(diff(dulieu[,i])))
 print("==========================================")
 
  
}
## [1] 5
## 
##  Phillips-Perron Unit Root Test
## 
## data:  dulieu[, i]
## Dickey-Fuller Z(alpha) = -86.91, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
## 
## 
##  Phillips-Perron Unit Root Test
## 
## data:  diff(dulieu[, i])
## Dickey-Fuller Z(alpha) = -105.9, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
## 
## [1] "=========================================="
## [1] 9
## 
##  Phillips-Perron Unit Root Test
## 
## data:  dulieu[, i]
## Dickey-Fuller Z(alpha) = -38.443, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
## 
## 
##  Phillips-Perron Unit Root Test
## 
## data:  diff(dulieu[, i])
## Dickey-Fuller Z(alpha) = -105.7, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
## 
## [1] "=========================================="
## [1] 11
## 
##  Phillips-Perron Unit Root Test
## 
## data:  dulieu[, i]
## Dickey-Fuller Z(alpha) = -23.16, Truncation lag parameter = 3, p-value
## = 0.02473
## alternative hypothesis: stationary
## 
## 
##  Phillips-Perron Unit Root Test
## 
## data:  diff(dulieu[, i])
## Dickey-Fuller Z(alpha) = -105.13, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
## 
## [1] "=========================================="
## [1] 13
## 
##  Phillips-Perron Unit Root Test
## 
## data:  dulieu[, i]
## Dickey-Fuller Z(alpha) = -18.383, Truncation lag parameter = 3, p-value
## = 0.08111
## alternative hypothesis: stationary
## 
## 
##  Phillips-Perron Unit Root Test
## 
## data:  diff(dulieu[, i])
## Dickey-Fuller Z(alpha) = -103.34, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
## 
## [1] "=========================================="
## [1] 14
## 
##  Phillips-Perron Unit Root Test
## 
## data:  dulieu[, i]
## Dickey-Fuller Z(alpha) = -14.522, Truncation lag parameter = 3, p-value
## = 0.2593
## alternative hypothesis: stationary
## 
## 
##  Phillips-Perron Unit Root Test
## 
## data:  diff(dulieu[, i])
## Dickey-Fuller Z(alpha) = -60.867, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
## 
## [1] "=========================================="
## [1] 18
## 
##  Phillips-Perron Unit Root Test
## 
## data:  dulieu[, i]
## Dickey-Fuller Z(alpha) = -7.2786, Truncation lag parameter = 3, p-value
## = 0.6892
## alternative hypothesis: stationary
## 
## 
##  Phillips-Perron Unit Root Test
## 
## data:  diff(dulieu[, i])
## Dickey-Fuller Z(alpha) = -70.804, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
## 
## [1] "=========================================="
## [1] 19
## 
##  Phillips-Perron Unit Root Test
## 
## data:  dulieu[, i]
## Dickey-Fuller Z(alpha) = -48.742, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
## 
## 
##  Phillips-Perron Unit Root Test
## 
## data:  diff(dulieu[, i])
## Dickey-Fuller Z(alpha) = -88.581, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
## 
## [1] "=========================================="
## [1] 20
## 
##  Phillips-Perron Unit Root Test
## 
## data:  dulieu[, i]
## Dickey-Fuller Z(alpha) = -2.3121, Truncation lag parameter = 3, p-value
## = 0.9582
## alternative hypothesis: stationary
## 
## 
##  Phillips-Perron Unit Root Test
## 
## data:  diff(dulieu[, i])
## Dickey-Fuller Z(alpha) = -90.39, Truncation lag parameter = 3, p-value
## = 0.01
## alternative hypothesis: stationary
## 
## [1] "=========================================="

Chạy ARDL

library(ARDL)
## To cite ARDL in publications use:
##   
## Kleanthis Natsiopoulos and Nickolaos Tzeremes (2020). ARDL: ARDL, ECM and Bounds-Test for Cointegration. R package version 0.1.0. University of Thessaly, Department of Economics. https://github.com/Natsiopoulos/ARDL.
congthuc1 <-LnGDP~ CPI + LnIG + LnIP + LnIF + LnLCU + LnLAB
congthuc2 <-CPI~ LnGDP + LnIG + LnIP + LnIF + LnLCU + LnLAB

nhieumodels <-auto_ardl(congthuc1, data=dulieu,max_order = 3)
nhieumodels
## $best_model
## 
## Time series regression with "ts" data:
## Start = 3, End = 80
## 
## Call:
## dynlm::dynlm(formula = full_formula, data = data, start = start, 
##     end = end)
## 
## Coefficients:
## (Intercept)  L(LnGDP, 1)  L(LnGDP, 2)          CPI    L(CPI, 1)         LnIG  
##   -1.859179     0.812249     0.220474     0.006563    -0.005649     0.451064  
##  L(LnIG, 1)         LnIP   L(LnIP, 1)   L(LnIP, 2)         LnIF   L(LnIF, 1)  
##   -0.392381     0.038740     0.065637    -0.158867     0.462531    -0.403662  
##  L(LnIF, 2)        LnLCU  L(LnLCU, 1)        LnLAB  L(LnLAB, 1)  L(LnLAB, 2)  
##   -0.066281     0.020045    -0.024492    -0.462715     1.213567    -0.615989  
## 
## 
## $best_order
## [1] 2 1 1 2 2 1 2
## 
## $top_orders
##    LnGDP CPI LnIG LnIP LnIF LnLCU LnLAB       AIC
## 1      2   1    1    2    2     1     2 -332.8700
## 2      2   1    1    2    2     1     1 -331.6097
## 3      2   1    1    2    2     2     2 -330.9760
## 4      2   1    1    2    1     1     2 -330.9391
## 5      2   2    1    2    2     1     2 -330.9366
## 6      2   1    2    2    2     1     2 -330.8900
## 7      2   1    2    2    2     2     2 -329.0064
## 8      2   1    1    2    1     2     2 -328.9625
## 9      2   1    1    3    2     1     2 -327.9577
## 10     2   2    2    2    2     2     2 -327.1070
## 11     2   1    1    3    2     2     2 -326.4064
## 12     1   1    1    0    1     1     2 -322.7270
## 13     1   1    1    0    1     1     1 -322.3274
## 14     1   1    1    0    2     1     2 -321.7530
## 15     1   2    1    0    1     1     2 -321.4645
## 16     1   1    1    1    1     1     1 -321.2375
## 17     1   1    2    0    1     1     2 -320.7935
## 18     1   1    1    0    1     2     2 -320.7315
## 19     1   1    1    0    1     0     1 -319.0068
## 20     1   1    1    0    2     1     1 -317.2323
modelchon1 <- nhieumodels$best_model
modelchon1
## 
## Time series regression with "ts" data:
## Start = 3, End = 80
## 
## Call:
## dynlm::dynlm(formula = full_formula, data = data, start = start, 
##     end = end)
## 
## Coefficients:
## (Intercept)  L(LnGDP, 1)  L(LnGDP, 2)          CPI    L(CPI, 1)         LnIG  
##   -1.859179     0.812249     0.220474     0.006563    -0.005649     0.451064  
##  L(LnIG, 1)         LnIP   L(LnIP, 1)   L(LnIP, 2)         LnIF   L(LnIF, 1)  
##   -0.392381     0.038740     0.065637    -0.158867     0.462531    -0.403662  
##  L(LnIF, 2)        LnLCU  L(LnLCU, 1)        LnLAB  L(LnLAB, 1)  L(LnLAB, 2)  
##   -0.066281     0.020045    -0.024492    -0.462715     1.213567    -0.615989
summary(modelchon1)
## 
## Time series regression with "ts" data:
## Start = 3, End = 80
## 
## Call:
## dynlm::dynlm(formula = full_formula, data = data, start = start, 
##     end = end)
## 
## Residuals:
##       Min        1Q    Median        3Q       Max 
## -0.094292 -0.012258 -0.001959  0.012030  0.070673 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)    
## (Intercept) -1.859179   2.180853  -0.853 0.397326    
## L(LnGDP, 1)  0.812249   0.080300  10.115 1.38e-14 ***
## L(LnGDP, 2)  0.220474   0.070175   3.142 0.002608 ** 
## CPI          0.006563   0.001326   4.950 6.35e-06 ***
## L(CPI, 1)   -0.005649   0.001259  -4.485 3.35e-05 ***
## LnIG         0.451064   0.040926  11.021 4.73e-16 ***
## L(LnIG, 1)  -0.392381   0.041227  -9.518 1.34e-13 ***
## LnIP         0.038740   0.044665   0.867 0.389207    
## L(LnIP, 1)   0.065637   0.053957   1.216 0.228575    
## L(LnIP, 2)  -0.158867   0.042571  -3.732 0.000424 ***
## LnIF         0.462531   0.022717  20.361  < 2e-16 ***
## L(LnIF, 1)  -0.403662   0.042851  -9.420 1.95e-13 ***
## L(LnIF, 2)  -0.066281   0.037637  -1.761 0.083328 .  
## LnLCU        0.020045   0.009796   2.046 0.045128 *  
## L(LnLCU, 1) -0.024492   0.009583  -2.556 0.013142 *  
## LnLAB       -0.462715   0.311036  -1.488 0.142078    
## L(LnLAB, 1)  1.213567   0.426254   2.847 0.006033 ** 
## L(LnLAB, 2) -0.615989   0.384913  -1.600 0.114778    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.0256 on 60 degrees of freedom
## Multiple R-squared:  0.9965, Adjusted R-squared:  0.9956 
## F-statistic:  1018 on 17 and 60 DF,  p-value: < 2.2e-16
bestmodel <-ardl(congthuc1,data=dulieu, order=c(4,4,1,1,4,2,4))
summary(bestmodel)
## 
## Time series regression with "ts" data:
## Start = 5, End = 80
## 
## Call:
## dynlm::dynlm(formula = full_formula, data = data, start = start, 
##     end = end)
## 
## Residuals:
##        Min         1Q     Median         3Q        Max 
## -0.0092300 -0.0024729 -0.0001102  0.0023951  0.0076504 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  6.627e-01  5.014e-01   1.322  0.19244    
## L(LnGDP, 1)  4.440e-02  2.366e-02   1.876  0.06658 .  
## L(LnGDP, 2) -3.288e-02  1.076e-02  -3.056  0.00362 ** 
## L(LnGDP, 3) -1.132e-02  1.029e-02  -1.101  0.27637    
## L(LnGDP, 4)  9.714e-01  2.380e-02  40.815  < 2e-16 ***
## CPI          4.400e-04  3.121e-04   1.410  0.16485    
## L(CPI, 1)   -9.196e-04  3.741e-04  -2.458  0.01754 *  
## L(CPI, 2)   -1.116e-05  3.191e-04  -0.035  0.97225    
## L(CPI, 3)   -6.201e-04  3.163e-04  -1.961  0.05562 .  
## L(CPI, 4)    3.734e-04  2.546e-04   1.467  0.14888    
## LnIG         1.364e-03  1.355e-02   0.101  0.92022    
## L(LnIG, 1)  -3.474e-02  1.119e-02  -3.105  0.00316 ** 
## LnIP         8.452e-03  8.612e-03   0.981  0.33118    
## L(LnIP, 1)   2.612e-02  9.132e-03   2.860  0.00620 ** 
## LnIF         1.911e-02  1.065e-02   1.794  0.07896 .  
## L(LnIF, 1)  -3.697e-02  1.159e-02  -3.190  0.00248 ** 
## L(LnIF, 2)  -2.187e-05  7.060e-03  -0.003  0.99754    
## L(LnIF, 3)   3.643e-03  6.958e-03   0.524  0.60295    
## L(LnIF, 4)   8.430e-03  5.233e-03   1.611  0.11365    
## LnLCU       -4.623e-05  1.947e-03  -0.024  0.98116    
## L(LnLCU, 1) -2.651e-03  2.099e-03  -1.263  0.21259    
## L(LnLCU, 2)  4.175e-03  2.043e-03   2.043  0.04641 *  
## LnLAB        5.110e-02  8.636e-02   0.592  0.55676    
## L(LnLAB, 1)  1.138e-01  8.440e-02   1.349  0.18360    
## L(LnLAB, 2)  3.386e-02  8.606e-02   0.393  0.69573    
## L(LnLAB, 3) -1.435e-02  8.031e-02  -0.179  0.85888    
## L(LnLAB, 4) -2.000e-01  8.611e-02  -2.323  0.02439 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.004502 on 49 degrees of freedom
## Multiple R-squared:  0.9999, Adjusted R-squared:  0.9999 
## F-statistic: 2.037e+04 on 26 and 49 DF,  p-value: < 2.2e-16

Ước lượng Dài hạn, ngắn hạn

lienket<-coint_eq(modelchon1,case=2)
lienket
## Time Series:
## Start = 1 
## End = 80 
## Frequency = 1 
##  [1] 13.36960 13.41629 13.38559 13.32778 13.10744 13.12933 13.06043 13.07099
##  [9] 13.16546 13.13817 13.07436 13.09664 13.50570 13.55481 13.54030 13.59693
## [17] 13.69841 13.61842 13.54386 13.58289 13.40572 13.37282 13.29115 13.30918
## [25] 13.43011 13.39028 13.29967 13.32455 13.75343 13.71192 13.63765 13.56908
## [33] 13.30319 13.02651 12.88210 12.98606 12.88160 12.91469 12.85178 12.80412
## [41] 13.50149 13.46624 13.48279 13.46815 13.19666 12.96001 12.74084 12.85129
## [49] 12.82026 12.99759 13.15321 13.25773 12.97174 13.10303 13.09170 13.27930
## [57] 13.35931 13.52249 13.61301 13.72445 13.47903 13.59096 13.57348 13.70566
## [65] 13.85794 13.86667 13.86877 13.92489 13.30949 13.17048 13.42025 13.40770
## [73] 13.50142 13.43375 13.56945 13.55240 13.63890 13.56991 13.80182 13.82779
shortrun <-uecm(modelchon1, case=2)
summary(shortrun)
## 
## Time series regression with "ts" data:
## Start = 3, End = 80
## 
## Call:
## dynlm::dynlm(formula = full_formula, data = data, start = start, 
##     end = end)
## 
## Residuals:
##       Min        1Q    Median        3Q       Max 
## -0.094292 -0.012258 -0.001959  0.012030  0.070673 
## 
## Coefficients:
##                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    -1.8591790  2.1808525  -0.853 0.397326    
## L(LnGDP, 1)     0.0327223  0.0727945   0.450 0.654679    
## L(CPI, 1)       0.0009137  0.0007278   1.256 0.214149    
## L(LnIG, 1)      0.0586827  0.0322004   1.822 0.073375 .  
## L(LnIP, 1)     -0.0544899  0.0300335  -1.814 0.074632 .  
## L(LnIF, 1)     -0.0074117  0.0326342  -0.227 0.821108    
## L(LnLCU, 1)    -0.0044471  0.0105109  -0.423 0.673742    
## L(LnLAB, 1)     0.1348633  0.2352470   0.573 0.568595    
## d(L(LnGDP, 1)) -0.2204736  0.0701749  -3.142 0.002608 ** 
## d(CPI)          0.0065626  0.0013258   4.950 6.35e-06 ***
## d(LnIG)         0.4510638  0.0409259  11.021 4.73e-16 ***
## d(LnIP)         0.0387402  0.0446650   0.867 0.389207    
## d(L(LnIP, 1))   0.1588669  0.0425711   3.732 0.000424 ***
## d(LnIF)         0.4625312  0.0227167  20.361  < 2e-16 ***
## d(L(LnIF, 1))   0.0662807  0.0376374   1.761 0.083328 .  
## d(LnLCU)        0.0200450  0.0097963   2.046 0.045128 *  
## d(LnLAB)       -0.4627149  0.3110361  -1.488 0.142078    
## d(L(LnLAB, 1))  0.6159887  0.3849128   1.600 0.114778    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.0256 on 60 degrees of freedom
## Multiple R-squared:  0.9927, Adjusted R-squared:  0.9906 
## F-statistic: 478.1 on 17 and 60 DF,  p-value: < 2.2e-16
bounds_f_test(modelchon1, case=3)
## 
##  Bounds F-test (Wald) for no cointegration
## 
## data:  d(LnGDP) ~ L(LnGDP, 1) + L(CPI, 1) + L(LnIG, 1) + L(LnIP, 1) +     L(LnIF, 1) + L(LnLCU, 1) + L(LnLAB, 1) + d(L(LnGDP, 1)) +     d(CPI) + d(LnIG) + d(LnIP) + d(L(LnIP, 1)) + d(LnIF) + d(L(LnIF,     1)) + d(LnLCU) + d(LnLAB) + d(L(LnLAB, 1))
## F = 2.0947, p-value = 0.4809
## alternative hypothesis: Possible cointegration
## null values:
##    k    T 
##    6 1000
bounds_t_test(modelchon1, case=3)
## 
##  Bounds t-test for no cointegration
## 
## data:  d(LnGDP) ~ L(LnGDP, 1) + L(CPI, 1) + L(LnIG, 1) + L(LnIP, 1) +     L(LnIF, 1) + L(LnLCU, 1) + L(LnLAB, 1) + d(L(LnGDP, 1)) +     d(CPI) + d(LnIG) + d(LnIP) + d(L(LnIP, 1)) + d(LnIF) + d(L(LnIF,     1)) + d(LnLCU) + d(LnLAB) + d(L(LnLAB, 1))
## t = 0.44952, p-value = 0.9936
## alternative hypothesis: Possible cointegration
## null values:
##    k    T 
##    6 1000

Chạy hết

ketqua1 <-multipliers(modelchon1, type = "lr")
ketqua1
##          term    estimate    std.error t.statistic   p.value
## 1 (Intercept) 56.81687311 120.16204397   0.4728354 0.6380462
## 2         CPI -0.02792359   0.05844355  -0.4777874 0.6345376
## 3        LnIG -1.79335368   4.29457604  -0.4175857 0.6777406
## 4        LnIP  1.66522076   3.20983971   0.5187863 0.6058180
## 5        LnIF  0.22650181   0.85410479   0.2651921 0.7917705
## 6       LnLCU  0.13590318   0.53490825   0.2540682 0.8003116
## 7       LnLAB -4.12144731  11.07047979  -0.3722917 0.7109875
ketqua2 <-multipliers(bestmodel)
ketqua2
##          term    estimate   std.error t.statistic   p.value
## 1 (Intercept) 23.30131743 32.55221229   0.7158136 0.4775039
## 2         CPI -0.02593033  0.02114944  -1.2260529 0.2260385
## 3        LnIG -1.17374772  1.24244457  -0.9447083 0.3494430
## 4        LnIP  1.21568977  1.08911281   1.1162202 0.2697719
## 5        LnIF -0.20454539  0.32866569  -0.6223509 0.5365966
## 6       LnLCU  0.05195888  0.13727640   0.3784983 0.7066955
## 7       LnLAB -0.54758204  2.77607222  -0.1972506 0.8444473
all.equal(ketqua1, ketqua2)
## [1] "Component \"estimate\": Mean relative difference: 0.5969625"  
## [2] "Component \"std.error\": Mean relative difference: 0.7278803" 
## [3] "Component \"t.statistic\": Mean relative difference: 1.188684"
## [4] "Component \"p.value\": Mean relative difference: 0.3530686"