# 1. 載入套件
library(readxl)
library(tidyverse)
Warning: package 'ggplot2' was built under R version 4.5.2
Warning: package 'tidyr' was built under R version 4.5.2
Warning: package 'purrr' was built under R version 4.5.2
Warning: package 'dplyr' was built under R version 4.5.2
Warning: package 'lubridate' was built under R version 4.5.2
── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr 1.2.0 ✔ readr 2.1.5
✔ forcats 1.0.1 ✔ stringr 1.5.2
✔ ggplot2 4.0.2 ✔ tibble 3.3.0
✔ lubridate 1.9.5 ✔ tidyr 1.3.2
✔ purrr 1.2.1
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag() masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
Attaching package: 'plm'
The following objects are masked from 'package:dplyr':
between, lag, lead
# 2. 讀取資料
hedging_data <- read_excel("~/Desktop/data/hedging.new.xlsx")
# 3. 指定成 df
df <- hedging_data
# 4. 修正欄名（全形底線改正常）
names(df) <- c(
"Country",
"Year",
"Political_Stability",
"Government_Effectiveness",
"Democracy",
"hedging",
"SCS",
"post2022",
"trade_China",
"sec_us"
)
# 5. 檢查欄名
names(df)
[1] "Country" "Year"
[3] "Political_Stability" "Government_Effectiveness"
[5] "Democracy" "hedging"
[7] "SCS" "post2022"
[9] "trade_China" "sec_us"
# ===============================
# ASEAN Hedging Panel Regression
# 你的研究正式版 R 程式
# ===============================
# 1. 載入套件
library(readxl)
library(tidyverse)
library(plm)
library(lmtest)
Loading required package: zoo
The following objects are masked from 'package:base':
as.Date, as.Date.numeric
library(sandwich)
# 2. 讀取資料
df <- read_excel("~/Desktop/data/hedging.new.xlsx")
# 3. 修正欄位名稱（避免全形底線）
names(df) <- c("Country",
"Year",
"Political_Stability",
"Government_Effectiveness",
"Democracy",
"hedging",
"SCS",
"post2022",
"trade_China",
"sec_us")
# 4. 資料型態轉換
df$Year <- as.numeric(df$Year)
df$hedging <- as.numeric(df$hedging)
df$Political_Stability <- as.numeric(df$Political_Stability)
df$Government_Effectiveness <- as.numeric(df$Government_Effectiveness)
df$Democracy <- as.numeric(df$Democracy)
df$SCS <- as.factor(df$SCS)
df$post2022 <- as.factor(df$post2022)
# 5. 建立 panel data
p_df <- pdata.frame(df, index = c("Country", "Year"))
# ===============================
# Model 1：固定效果模型（主模型）
# ===============================
model1 <- plm(
hedging ~ Political_Stability +
Government_Effectiveness +
Democracy +
post2022,
data = p_df,
model = "within"
)
summary(model1)
Oneway (individual) effect Within Model
Call:
plm(formula = hedging ~ Political_Stability + Government_Effectiveness +
Democracy + post2022, data = p_df, model = "within")
Balanced Panel: n = 8, T = 10, N = 80
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-0.0699649 -0.0063760 0.0018528 0.0091569 0.0434486
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
Political_Stability 0.00097050 0.00097604 0.9943 0.323593
Government_Effectiveness -0.00070885 0.00109552 -0.6470 0.519781
Democracy -0.00694831 0.00577214 -1.2038 0.232853
post20221 0.02030061 0.00624377 3.2513 0.001789 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 0.035588
Residual Sum of Squares: 0.027011
R-Squared: 0.24101
Adj. R-Squared: 0.11823
F-statistic: 5.39816 on 4 and 68 DF, p-value: 0.00077904
# robust SE
coeftest(model1, vcov = vcovHC(model1, type = "HC1"))
t test of coefficients:
Estimate Std. Error t value Pr(>|t|)
Political_Stability 0.00097050 0.00089709 1.0818 0.283153
Government_Effectiveness -0.00070885 0.00061144 -1.1593 0.250382
Democracy -0.00694831 0.01038215 -0.6693 0.505598
post20221 0.02030061 0.00708538 2.8651 0.005541 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# ===============================
# Model 2：加入南海變數（Random Effects）
# 固定效果會吸收不變變數SCS
# ===============================
model2 <- plm(
hedging ~ Political_Stability +
Government_Effectiveness +
Democracy +
SCS +
post2022,
data = p_df,
model = "random"
)
summary(model2)
Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = hedging ~ Political_Stability + Government_Effectiveness +
Democracy + SCS + post2022, data = p_df, model = "random")
Balanced Panel: n = 8, T = 10, N = 80
Effects:
var std.dev share
idiosyncratic 0.0003972 0.0199304 0.04
individual 0.0094434 0.0971773 0.96
theta: 0.9353
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-0.06726651 -0.00924393 -0.00066896 0.00992457 0.04707561
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 0.11708782 0.09202750 1.2723 0.2032618
Political_Stability 0.00094776 0.00091910 1.0312 0.3024574
Government_Effectiveness -0.00040324 0.00097798 -0.4123 0.6801060
Democracy -0.00650175 0.00551423 -1.1791 0.2383645
SCS1 0.01799247 0.06887228 0.2612 0.7939044
post20221 0.02074811 0.00610092 3.4008 0.0006719 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 0.036952
Residual Sum of Squares: 0.02851
R-Squared: 0.22845
Adj. R-Squared: 0.17632
Chisq: 21.9105 on 5 DF, p-value: 0.00054452
coeftest(model2, vcov = vcovHC(model2, type = "HC1"))
t test of coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.11708782 0.07235560 1.6182 0.109868
Political_Stability 0.00094776 0.00083338 1.1372 0.259104
Government_Effectiveness -0.00040324 0.00053086 -0.7596 0.449911
Democracy -0.00650175 0.01002102 -0.6488 0.518469
SCS1 0.01799247 0.05130673 0.3507 0.726821
post20221 0.02074811 0.00729695 2.8434 0.005767 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# ===============================
# Model 3：治理能力 × post2022
# 國際壓力下治理國家是否更能避險
# ===============================
model3 <- plm(
hedging ~ Government_Effectiveness * post2022 +
Political_Stability +
Democracy,
data = p_df,
model = "within"
)
summary(model3)
Oneway (individual) effect Within Model
Call:
plm(formula = hedging ~ Government_Effectiveness * post2022 +
Political_Stability + Democracy, data = p_df, model = "within")
Balanced Panel: n = 8, T = 10, N = 80
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-0.0700787 -0.0062442 0.0018180 0.0091225 0.0435024
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
Government_Effectiveness -6.9831e-04 1.1240e-03 -0.6213 0.5365
post20221 2.1366e-02 2.2443e-02 0.9520 0.3445
Political_Stability 9.5543e-04 1.0294e-03 0.9281 0.3567
Democracy -6.8860e-03 5.9502e-03 -1.1573 0.2513
Government_Effectiveness:post20221 -1.7848e-05 3.6104e-04 -0.0494 0.9607
Total Sum of Squares: 0.035588
Residual Sum of Squares: 0.02701
R-Squared: 0.24104
Adj. R-Squared: 0.1051
F-statistic: 4.25567 on 5 and 67 DF, p-value: 0.0020241
coeftest(model3, vcov = vcovHC(model3, type = "HC1"))
t test of coefficients:
Estimate Std. Error t value Pr(>|t|)
Government_Effectiveness -6.9831e-04 7.3178e-04 -0.9543 0.3434
post20221 2.1366e-02 2.1175e-02 1.0090 0.3166
Political_Stability 9.5543e-04 9.3099e-04 1.0263 0.3085
Democracy -6.8860e-03 9.9446e-03 -0.6924 0.4911
Government_Effectiveness:post20221 -1.7848e-05 3.0354e-04 -0.0588 0.9533
# ===============================
# Hausman Test
# Fixed vs Random
# ===============================
phtest(model1, model2)
Hausman Test
data: hedging ~ Political_Stability + Government_Effectiveness + Democracy + ...
chisq = 0.47959, df = 4, p-value = 0.9755
alternative hypothesis: one model is inconsistent
# ===============================
# VIF（共線性檢查）
# ===============================
library(car)
Loading required package: carData
The following object is masked from 'package:dplyr':
recode
The following object is masked from 'package:purrr':
some
ols_model <- lm(
hedging ~ Political_Stability +
Government_Effectiveness +
Democracy +
post2022,
data = df
)
vif(ols_model)
Political_Stability Government_Effectiveness Democracy
4.193471 5.805257 4.637602
post2022
1.038000
# ===============================
# 結果輸出（可做表格）
# ===============================
library(stargazer)
Hlavac, Marek (2022). stargazer: Well-Formatted Regression and Summary Statistics Tables.
R package version 5.2.3. https://CRAN.R-project.org/package=stargazer
stargazer(model1, model2, model3,
type = "text",
title = "Determinants of ASEAN Hedging",
digits = 3)
Determinants of ASEAN Hedging
========================================================================================
Dependent variable:
-----------------------------------------------------
hedging
(1) (2) (3)
----------------------------------------------------------------------------------------
Political_Stability 0.001 0.001 0.001
(0.001) (0.001) (0.001)
Government_Effectiveness -0.001 -0.0004 -0.001
(0.001) (0.001) (0.001)
Democracy -0.007 -0.007 -0.007
(0.006) (0.006) (0.006)
SCS1 0.018
(0.069)
Government_Effectiveness:post20221 -0.00002
(0.0004)
post20221 0.020*** 0.021*** 0.021
(0.006) (0.006) (0.022)
Constant 0.117
(0.092)
----------------------------------------------------------------------------------------
Observations 80 80 80
R2 0.241 0.228 0.241
Adjusted R2 0.118 0.176 0.105
F Statistic 5.398*** (df = 4; 68) 21.910*** 4.256*** (df = 5; 67)
========================================================================================
Note: *p<0.1; **p<0.05; ***p<0.01
