library(readxl)
library(vars)
## Warning: package 'vars' was built under R version 4.6.1
## Loading required package: MASS
## Loading required package: strucchange
## Warning: package 'strucchange' was built under R version 4.6.1
## Loading required package: zoo
##
## Attaching package: 'zoo'
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric
## Loading required package: sandwich
## Warning: package 'sandwich' was built under R version 4.6.1
## Loading required package: urca
## Loading required package: lmtest
library(urca)
# 1. IMPORT DATA & PERSIAPAN OBJEK TIME SERIES
dataVAR <- read_excel("C:/Dhinar/ARW/Data No. 2 ARW UAS.xlsx")
dataVAR <- data.frame(dataVAR)
# Standarisasi nama kolom agar tidak merusak fungsi VAR/IRF
colnames(dataVAR) <- make.names(colnames(dataVAR))
# Membentuk time series
tsdataVAR <- ts(dataVAR[, c("Inflasi", "BI..Rate", "USD")], start = c(2011, 1), frequency = 12)
# Definisi data_pakai (MEMPERBAIKI ERROR 'data_pakai not found')
data_pakai <- tsdataVAR
# Plot deret waktu
plot(data_pakai, main = "Plot Waktu: Inflasi, BI-Rate, dan USD/IDR")
# 2. UJI STASIONERITAS (ADF TEST - LEVEL)
inf <- data_pakai[, "Inflasi"]
bir <- data_pakai[, "BI..Rate"]
usd <- data_pakai[, "USD"]
adf_inf_drift <- ur.df(inf, type = "drift", selectlags = "AIC")
summary(adf_inf_drift)
##
## ###############################################
## # Augmented Dickey-Fuller Test Unit Root Test #
## ###############################################
##
## Test regression drift
##
##
## Call:
## lm(formula = z.diff ~ z.lag.1 + 1 + z.diff.lag)
##
## Residuals:
## Min 1Q Median 3Q Max
## -64.706 -16.229 0.107 17.924 68.822
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 43.18817 4.44546 9.715 < 2e-16 ***
## z.lag.1 -0.91081 0.08517 -10.694 < 2e-16 ***
## z.diff.lag 0.27106 0.07094 3.821 0.000182 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 25.23 on 183 degrees of freedom
## Multiple R-squared: 0.4055, Adjusted R-squared: 0.399
## F-statistic: 62.41 on 2 and 183 DF, p-value: < 2.2e-16
##
##
## Value of test-statistic is: -10.6936 57.1772
##
## Critical values for test statistics:
## 1pct 5pct 10pct
## tau2 -3.46 -2.88 -2.57
## phi1 6.52 4.63 3.81
adf_bir_drift <- ur.df(bir, type = "drift", selectlags = "AIC")
summary(adf_bir_drift)
##
## ###############################################
## # Augmented Dickey-Fuller Test Unit Root Test #
## ###############################################
##
## Test regression drift
##
##
## Call:
## lm(formula = z.diff ~ z.lag.1 + 1 + z.diff.lag)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.9169 -0.1438 0.0264 0.1588 1.9508
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.162680 0.105133 1.547 0.1235
## z.lag.1 -0.018909 0.009968 -1.897 0.0594 .
## z.diff.lag 0.465032 0.064863 7.169 1.8e-11 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.663 on 183 degrees of freedom
## Multiple R-squared: 0.2259, Adjusted R-squared: 0.2175
## F-statistic: 26.7 on 2 and 183 DF, p-value: 6.671e-11
##
##
## Value of test-statistic is: -1.8969 1.8415
##
## Critical values for test statistics:
## 1pct 5pct 10pct
## tau2 -3.46 -2.88 -2.57
## phi1 6.52 4.63 3.81
adf_usd_drift <- ur.df(usd, type = "drift", selectlags = "AIC")
summary(adf_usd_drift)
##
## ###############################################
## # Augmented Dickey-Fuller Test Unit Root Test #
## ###############################################
##
## Test regression drift
##
##
## Call:
## lm(formula = z.diff ~ z.lag.1 + 1 + z.diff.lag)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1063.70 -124.79 -16.86 155.68 2133.65
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 215.50202 143.39140 1.503 0.135
## z.lag.1 -0.01210 0.01044 -1.159 0.248
## z.diff.lag -0.07677 0.07378 -1.040 0.299
##
## Residual standard error: 328.5 on 183 degrees of freedom
## Multiple R-squared: 0.01409, Adjusted R-squared: 0.003315
## F-statistic: 1.308 on 2 and 183 DF, p-value: 0.273
##
##
## Value of test-statistic is: -1.1587 2.9312
##
## Critical values for test statistics:
## 1pct 5pct 10pct
## tau2 -3.46 -2.88 -2.57
## phi1 6.52 4.63 3.81
# 3. PENENTUAN LAG OPTIMUM VAR
lag_opt <- VARselect(data_pakai, lag.max = 10, type = "const")
print(lag_opt$selection)
## AIC(n) HQ(n) SC(n) FPE(n)
## 4 2 2 4
# Memilih lag optimum berdasarkan kriteria AIC
p_opt <- lag_opt$selection["AIC(n)"]
# 4. PEMODELAN VAR LEVEL
model_var <- VAR(data_pakai, p = p_opt, type = "const")
summary(model_var)
##
## VAR Estimation Results:
## =========================
## Endogenous variables: Inflasi, BI..Rate, USD
## Deterministic variables: const
## Sample size: 184
## Log Likelihood: -2325.887
## Roots of the characteristic polynomial:
## 0.9894 0.9168 0.755 0.755 0.734 0.5916 0.5916 0.5262 0.5262 0.4796 0.3803 0.3798
## Call:
## VAR(y = data_pakai, p = p_opt, type = "const")
##
##
## Estimation results for equation Inflasi:
## ========================================
## Inflasi = Inflasi.l1 + BI..Rate.l1 + USD.l1 + Inflasi.l2 + BI..Rate.l2 + USD.l2 + Inflasi.l3 + BI..Rate.l3 + USD.l3 + Inflasi.l4 + BI..Rate.l4 + USD.l4 + const
##
## Estimate Std. Error t value Pr(>|t|)
## Inflasi.l1 0.226182 0.077219 2.929 0.00386 **
## BI..Rate.l1 2.257679 2.988937 0.755 0.45108
## USD.l1 0.003148 0.005790 0.544 0.58736
## Inflasi.l2 -0.259354 0.078793 -3.292 0.00121 **
## BI..Rate.l2 -2.306287 4.881780 -0.472 0.63722
## USD.l2 0.003769 0.007540 0.500 0.61780
## Inflasi.l3 -0.124059 0.079299 -1.564 0.11956
## BI..Rate.l3 -2.355827 4.924690 -0.478 0.63300
## USD.l3 -0.002770 0.007542 -0.367 0.71392
## Inflasi.l4 -0.137577 0.075984 -1.811 0.07196 .
## BI..Rate.l4 2.575653 2.942114 0.875 0.38256
## USD.l4 -0.006980 0.005877 -1.188 0.23664
## const 97.406336 17.051673 5.712 4.84e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
##
## Residual standard error: 24.61 on 171 degrees of freedom
## Multiple R-Squared: 0.2339, Adjusted R-squared: 0.1801
## F-statistic: 4.35 on 12 and 171 DF, p-value: 4.958e-06
##
##
## Estimation results for equation BI..Rate:
## =========================================
## BI..Rate = Inflasi.l1 + BI..Rate.l1 + USD.l1 + Inflasi.l2 + BI..Rate.l2 + USD.l2 + Inflasi.l3 + BI..Rate.l3 + USD.l3 + Inflasi.l4 + BI..Rate.l4 + USD.l4 + const
##
## Estimate Std. Error t value Pr(>|t|)
## Inflasi.l1 2.509e-04 2.013e-03 0.125 0.9010
## BI..Rate.l1 1.318e+00 7.793e-02 16.914 <2e-16 ***
## USD.l1 2.708e-04 1.510e-04 1.794 0.0746 .
## Inflasi.l2 -3.704e-03 2.054e-03 -1.803 0.0731 .
## BI..Rate.l2 -1.630e-01 1.273e-01 -1.281 0.2021
## USD.l2 3.561e-05 1.966e-04 0.181 0.8565
## Inflasi.l3 -9.880e-04 2.068e-03 -0.478 0.6334
## BI..Rate.l3 -1.379e-01 1.284e-01 -1.074 0.2843
## USD.l3 -3.000e-04 1.967e-04 -1.526 0.1289
## Inflasi.l4 3.781e-03 1.981e-03 1.909 0.0580 .
## BI..Rate.l4 -4.439e-02 7.671e-02 -0.579 0.5635
## USD.l4 -2.812e-05 1.532e-04 -0.184 0.8546
## const 5.376e-01 4.446e-01 1.209 0.2282
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
##
## Residual standard error: 0.6416 on 171 degrees of freedom
## Multiple R-Squared: 0.9839, Adjusted R-squared: 0.9828
## F-statistic: 872.2 on 12 and 171 DF, p-value: < 2.2e-16
##
##
## Estimation results for equation USD:
## ====================================
## USD = Inflasi.l1 + BI..Rate.l1 + USD.l1 + Inflasi.l2 + BI..Rate.l2 + USD.l2 + Inflasi.l3 + BI..Rate.l3 + USD.l3 + Inflasi.l4 + BI..Rate.l4 + USD.l4 + const
##
## Estimate Std. Error t value Pr(>|t|)
## Inflasi.l1 0.89353 1.00989 0.885 0.3775
## BI..Rate.l1 86.65131 39.09003 2.217 0.0280 *
## USD.l1 0.82988 0.07573 10.959 <2e-16 ***
## Inflasi.l2 0.30806 1.03047 0.299 0.7653
## BI..Rate.l2 -64.58268 63.84508 -1.012 0.3132
## USD.l2 -0.03392 0.09861 -0.344 0.7313
## Inflasi.l3 2.41646 1.03709 2.330 0.0210 *
## BI..Rate.l3 -83.21548 64.40627 -1.292 0.1981
## USD.l3 0.04715 0.09864 0.478 0.6333
## Inflasi.l4 0.39089 0.99373 0.393 0.6945
## BI..Rate.l4 61.16653 38.47767 1.590 0.1138
## USD.l4 0.15056 0.07686 1.959 0.0518 .
## const -28.51611 223.00584 -0.128 0.8984
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
##
## Residual standard error: 321.8 on 171 degrees of freedom
## Multiple R-Squared: 0.9811, Adjusted R-squared: 0.9798
## F-statistic: 741.6 on 12 and 171 DF, p-value: < 2.2e-16
##
##
##
## Covariance matrix of residuals:
## Inflasi BI..Rate USD
## Inflasi 605.495 3.8329 450.86
## BI..Rate 3.833 0.4116 27.07
## USD 450.862 27.0659 103564.09
##
## Correlation matrix of residuals:
## Inflasi BI..Rate USD
## Inflasi 1.00000 0.2428 0.05694
## BI..Rate 0.24279 1.0000 0.13109
## USD 0.05694 0.1311 1.00000
# 5. IMPULSE RESPONSE FUNCTION (IRF) DAN PLOT
irf_inf_bir <- irf(model_var, impulse = "Inflasi", response = "BI..Rate", n.ahead = 12, boot = TRUE)
irf_inf_usd <- irf(model_var, impulse = "Inflasi", response = "USD", n.ahead = 12, boot = TRUE)
irf_bir_inf <- irf(model_var, impulse = "BI..Rate", response = "Inflasi", n.ahead = 12, boot = TRUE)
irf_usd_inf <- irf(model_var, impulse = "USD", response = "Inflasi", n.ahead = 12, boot = TRUE)
# Reset device grafik jika terjadi error "pin" / margin
if(!is.null(dev.list())) dev.off()
## null device
## 1
# Tampilkan grafik IRF dalam satu layout 2x2
par(mfrow = c(2, 2))
plot(irf_inf_bir, main = "Shock Inflasi -> BI-Rate")
plot(irf_inf_usd, main = "Shock Inflasi -> USD")
plot(irf_bir_inf, main = "Shock BI-Rate -> Inflasi")
plot(irf_usd_inf, main = "Shock USD -> Inflasi")