library(readxl)
library(dplyr)
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
library(plm)
##
## Attaching package: 'plm'
## The following objects are masked from 'package:dplyr':
##
## between, lag, lead
library(lmtest)
## Loading required package: zoo
##
## Attaching package: 'zoo'
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric
# ===== 1. Import =====
path <- "C:/Users/ADVAN/OneDrive/Dokumen/Data_Panel_WDI.xlsx"
data <- read_excel(path)
# Rapikan nama kolom (spasi/simbol -> titik)
names(data) <- make.names(names(data))
# Lihat nama kolom & isi data
print(names(data))
## [1] "Negara" "Kode"
## [3] "Tahun" "Y_Kemiskinan"
## [5] "X1_Melek_Huruf" "X2_TPT"
## [7] "X3_PDRB_per_kapita" "X4_Pengeluaran_Pemerintah"
head(data)
## # A tibble: 6 × 8
## Negara Kode Tahun Y_Kemiskinan X1_Melek_Huruf X2_TPT X3_PDRB_per_kapita
## <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 Australia AUS 2000 NA NA 6.29 2.73
## 2 Australia AUS 2001 NA NA 6.75 0.730
## 3 Australia AUS 2002 NA NA 6.38 2.77
## 4 Australia AUS 2003 NA NA 5.93 1.91
## 5 Australia AUS 2004 NA NA 5.40 3.11
## 6 Australia AUS 2005 NA NA 5.04 1.91
## # ℹ 1 more variable: X4_Pengeluaran_Pemerintah <dbl>
# ===== 2. Deteksi otomatis kolom negara & tahun =====
id_col <- names(data)[grep("country|negara|economy|code", names(data), ignore.case = TRUE)][1]
time_col <- names(data)[grep("year|tahun|time", names(data), ignore.case = TRUE)][1]
cat("Kolom individu:", id_col, "\n")
## Kolom individu: Negara
cat("Kolom waktu :", time_col, "\n")
## Kolom waktu : Tahun
# Pastikan tahun berupa angka
data[[time_col]] <- as.numeric(as.character(data[[time_col]]))
# ===== 3. Struktur panel =====
pdata <- pdata.frame(data, index = c(id_col, time_col))
pdim(pdata)
## Balanced Panel: n = 15, T = 26, N = 390
# ===== 4. Pilih variabel =====
# Semua kolom numerik selain tahun
num_vars <- setdiff(names(data)[sapply(data, is.numeric)], time_col)
print(num_vars)
## [1] "Y_Kemiskinan" "X1_Melek_Huruf"
## [3] "X2_TPT" "X3_PDRB_per_kapita"
## [5] "X4_Pengeluaran_Pemerintah"
# GANTI sesuai variabel Anda, contoh: Y = num_vars[1], X = sisanya
y_var <- num_vars[1]
x_vars <- num_vars[-1]
rumus <- as.formula(paste(y_var, "~", paste(x_vars, collapse = " + ")))
print(rumus)
## Y_Kemiskinan ~ X1_Melek_Huruf + X2_TPT + X3_PDRB_per_kapita +
## X4_Pengeluaran_Pemerintah
# ===== 5. Model =====
model_pooled <- plm(rumus, data = pdata, model = "pooling")
model_fe <- plm(rumus, data = pdata, model = "within")
model_re <- plm(rumus, data = pdata, model = "random")
summary(model_pooled)
## Pooling Model
##
## Call:
## plm(formula = rumus, data = pdata, model = "pooling")
##
## Unbalanced Panel: n = 7, T = 1-11, N = 29
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -12.320 -4.116 -1.960 0.517 29.679
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## (Intercept) 98.21491 33.77095 2.9083 0.007709 **
## X1_Melek_Huruf -0.95485 0.34468 -2.7702 0.010639 *
## X2_TPT 0.34634 0.84721 0.4088 0.686313
## X3_PDRB_per_kapita 1.46959 0.81971 1.7928 0.085616 .
## X4_Pengeluaran_Pemerintah -0.20441 0.16260 -1.2571 0.220816
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 4948.2
## Residual Sum of Squares: 2470.1
## R-Squared: 0.5008
## Adj. R-Squared: 0.41761
## F-statistic: 6.01934 on 4 and 24 DF, p-value: 0.0016786
summary(model_fe)
## Oneway (individual) effect Within Model
##
## Call:
## plm(formula = rumus, data = pdata, model = "within")
##
## Unbalanced Panel: n = 7, T = 1-11, N = 29
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -10.549 -4.598 -0.387 1.812 26.017
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## X1_Melek_Huruf -2.74859 1.28124 -2.1453 0.04582 *
## X2_TPT -1.66568 2.37957 -0.7000 0.49288
## X3_PDRB_per_kapita 1.18560 0.95697 1.2389 0.23129
## X4_Pengeluaran_Pemerintah -1.72908 0.69452 -2.4896 0.02279 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 2547.1
## Residual Sum of Squares: 1610.2
## R-Squared: 0.36784
## Adj. R-Squared: 0.016644
## F-statistic: 2.61848 on 4 and 18 DF, p-value: 0.069496
summary(model_re)
## Oneway (individual) effect Random Effect Model
## (Swamy-Arora's transformation)
##
## Call:
## plm(formula = rumus, data = pdata, model = "random")
##
## Unbalanced Panel: n = 7, T = 1-11, N = 29
##
## Effects:
## var std.dev share
## idiosyncratic 89.453 9.458 0.688
## individual 40.474 6.362 0.312
## theta:
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.1702 0.3487 0.5347 0.4706 0.5910 0.5910
##
## Residuals:
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -8.5820 -4.9507 -1.8171 0.1120 0.0365 28.4400
##
## Coefficients:
## Estimate Std. Error z-value Pr(>|z|)
## (Intercept) 112.57584 43.34531 2.5972 0.009399 **
## X1_Melek_Huruf -1.11903 0.44619 -2.5080 0.012143 *
## X2_TPT 0.77709 1.35101 0.5752 0.565160
## X3_PDRB_per_kapita 1.52137 0.86664 1.7555 0.079177 .
## X4_Pengeluaran_Pemerintah -0.29147 0.20886 -1.3955 0.162866
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 4201.7
## Residual Sum of Squares: 2229.1
## R-Squared: 0.46968
## Adj. R-Squared: 0.38129
## Chisq: 15.7647 on 4 DF, p-value: 0.0033517
# ===== 6. Pemilihan model =====
pFtest(model_fe, model_pooled) # Chow
##
## F test for individual effects
##
## data: rumus
## F = 1.6023, df1 = 6, df2 = 18, p-value = 0.2037
## alternative hypothesis: significant effects
phtest(model_fe, model_re) # Hausman
##
## Hausman Test
##
## data: rumus
## chisq = 5.1868, df = 4, p-value = 0.2687
## alternative hypothesis: one model is inconsistent
plmtest(model_pooled, type = "bp") # Breusch-Pagan LM
##
## Lagrange Multiplier Test - (Breusch-Pagan)
##
## data: rumus
## chisq = 0.60375, df = 1, p-value = 0.4372
## alternative hypothesis: significant effects
# ===== 7. Uji asumsi & robust SE =====
bptest(rumus, data = data)
##
## studentized Breusch-Pagan test
##
## data: rumus
## BP = 1.3407, df = 4, p-value = 0.8544
pbgtest(model_fe)
##
## Breusch-Godfrey/Wooldridge test for serial correlation in panel models
##
## data: rumus
## chisq = 0.82508, df = 1, p-value = 0.3637
## alternative hypothesis: serial correlation in idiosyncratic errors
coeftest(model_fe, vcov = vcovHC(model_fe, type = "HC1", cluster = "group"))
##
## t test of coefficients:
##
## Estimate Std. Error t value Pr(>|t|)
## X1_Melek_Huruf -2.74859 2.31199 -1.1888 0.24995
## X2_TPT -1.66568 4.32172 -0.3854 0.70444
## X3_PDRB_per_kapita 1.18560 0.89769 1.3207 0.20314
## X4_Pengeluaran_Pemerintah -1.72908 0.90919 -1.9018 0.07333 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.
When you click the Knit button a document will be generated that includes both content as well as the output of any embedded R code chunks within the document. You can embed an R code chunk like this:
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. :120.00
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.