# 1. 載入核心套件 (避開 ggstatsplot 以免版本報錯) 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 <- read.csv ("aseananalysis.csv" )# 3. 確保數值型態正確 $ Align <- as.numeric (df$ Align)$ SCS <- as.factor (df$ SCS) # 0 = 無爭議, 1 = 有爭議 # 4. 建立面板數據格式 (指定國家與年份) <- pdata.frame (df, index = c ("Country" , "Year" ))
四分位趨勢分析 (Boxplot Analysis)
用來觀察東協整體的投票分布，看看 2024 年補值後，整體重心是否上移。
# 繪製各年份的四分位圖 ggplot (df, aes (x = factor (Year), y = Align)) + geom_boxplot (aes (fill = factor (Year)), alpha = 0.5 , outlier.color = "red" ) + geom_jitter (width = 0.2 , alpha = 0.4 , color = "black" ) + # 顯示每個國家的點 scale_fill_brewer (palette = "Set3" ) + theme_minimal () + labs (title = "ASEAN Alignment Score Distribution (2014-2024)" ,subtitle = "Higher Score = More Pro-US | Lower Score = More Pro-China" ,x = "Year" , y = "Alignment Score" ) + theme (legend.position = "none" )
南海爭議與投票傾向 (Chi-Square Test)
我們要測試「是否有南海爭議」與「投票傾向（親美/親中）」是否有顯著關聯。
# A. 將 Align 轉換為類別：高於中位數為 "Pro-US", 低於則為 "Pro-China" <- df %>% mutate (Align_Type = ifelse (Align > median (Align, na.rm= T), "Pro-US" , "Pro-China" ))# B. 執行卡方檢定 (SCS vs Align_Type) <- table (df_cat$ SCS, df_cat$ Align_Type)print ("--- 交叉表 (0=無爭議, 1=有爭議) ---" )
[1] "--- 交叉表 (0=無爭議, 1=有爭議) ---"
Pro-China Pro-US
0 20 24
1 24 20
<- chisq.test (contingency_table)print ("--- 卡方檢定結果 ---" )
Pearson's Chi-squared test with Yates' continuity correction
data: contingency_table
X-squared = 0.40909, df = 1, p-value = 0.5224
# C. 視覺化關聯 (用百分比堆疊條形圖代替 ggstatsplot) ggplot (df_cat, aes (x = SCS, fill = Align_Type)) + geom_bar (position = "fill" ) + scale_y_continuous (labels = scales:: percent) + theme_minimal () + labs (title = "Impact of SCS Dispute on Voting Alignment" ,x = "SCS Dispute (0 = No, 1 = Yes)" ,y = "Percentage" , fill = "Alignment Group" )
因果推論回歸 (Panel Regression)
利用固定效應模型（FixedEffects），分析在排除國家固定背景後，貿易與軍事變數的影響。
# 執行固定效應模型 (Within-model) # 探討：對中/美貿易、中/美軍援、民主度、南海爭議對投票的影響 <- plm (Align ~ Trade_Chn_GDP + Trade_US_GDP + + Mil_US + + SCS, data = p_df, model = "within" )# 顯示統計結果 print ("--- 固定效應回歸報告 ---" )
Oneway (individual) effect Within Model
Call:
plm(formula = Align ~ Trade_Chn_GDP + Trade_US_GDP + Mil_China +
Mil_US + Democracy + SCS, data = p_df, model = "within")
Balanced Panel: n = 8, T = 11, N = 88
Residuals:
Cambodia-2014 Cambodia-2015 Cambodia-2016 Cambodia-2017
-9.9077e-16 -9.9655e-16 -9.9344e-16 -9.9531e-16
Cambodia-2018 Cambodia-2019 Cambodia-2020 Cambodia-2021
-4.4816e-02 4.4816e-02 2.8081e-03 -9.8684e-16
Cambodia-2022 Cambodia-2023 Cambodia-2024 Indonesia-2014
-3.6635e-03 8.5538e-04 -9.9285e-16 -1.5635e-15
Indonesia-2015 Indonesia-2016 Indonesia-2017 Indonesia-2018
-1.5366e-15 -1.5302e-15 -1.5646e-15 -1.5561e-15
Indonesia-2019 Indonesia-2020 Indonesia-2021 Indonesia-2022
-1.6133e-15 -1.5926e-15 -1.5985e-15 -1.5871e-15
Indonesia-2023 Indonesia-2024 Lao PDR-2014 Lao PDR-2015
-1.5498e-15 -1.5669e-15 -1.2171e-01 1.9531e-01
Lao PDR-2016 Lao PDR-2017 Lao PDR-2018 Lao PDR-2019
1.0720e-02 -1.9359e-02 -6.4863e-02 -1.0180e-04
Lao PDR-2020 Lao PDR-2021 Lao PDR-2022 Lao PDR-2023
-1.7539e-15 -1.7103e-15 -3.6635e-03 3.6635e-03
Lao PDR-2024 Malaysia-2014 Malaysia-2015 Malaysia-2016
-1.7474e-15 -1.6657e-15 -1.6631e-15 -1.6624e-15
Malaysia-2017 Malaysia-2018 Malaysia-2019 Malaysia-2020
-1.6539e-15 -1.6627e-15 -1.6594e-15 -1.6644e-15
Malaysia-2021 Malaysia-2022 Malaysia-2023 Malaysia-2024
-1.6686e-15 -1.6612e-15 -1.6495e-15 -1.6832e-15
Philippines-2014 Philippines-2015 Philippines-2016 Philippines-2017
-5.5444e-16 -5.6430e-16 -5.6085e-16 -5.6398e-16
Philippines-2018 Philippines-2019 Philippines-2020 Philippines-2021
-5.6841e-16 -5.7078e-16 -5.6450e-16 -5.6073e-16
Philippines-2022 Philippines-2023 Philippines-2024 Singapore-2014
-5.5662e-16 -5.6633e-16 -5.5470e-16 -3.5201e-16
Singapore-2015 Singapore-2016 Singapore-2017 Singapore-2018
-4.5593e-16 -4.5780e-16 -4.5692e-16 -4.5903e-16
Singapore-2019 Singapore-2020 Singapore-2021 Singapore-2022
-5.2144e-16 -4.5696e-16 -4.3664e-16 -5.7021e-16
Singapore-2023 Singapore-2024 Thailand-2014 Thailand-2015
-4.1929e-16 -4.3765e-16 3.7411e-17 3.6626e-17
Thailand-2016 Thailand-2017 Thailand-2018 Thailand-2019
3.4786e-17 4.9894e-17 3.2729e-17 3.0032e-17
Thailand-2020 Thailand-2021 Thailand-2022 Thailand-2023
3.0957e-17 2.9302e-17 3.5041e-17 3.5403e-17
Thailand-2024 Vietnam-2014 Vietnam-2015 Vietnam-2016
4.9331e-17 1.5268e-15 1.5468e-15 1.5396e-15
Vietnam-2017 Vietnam-2018 Vietnam-2019 Vietnam-2020
1.5304e-15 1.5384e-15 1.5380e-15 1.5380e-15
Vietnam-2021 Vietnam-2022 Vietnam-2023 Vietnam-2024
1.5372e-15 1.5428e-15 1.5408e-15 1.5402e-15
Coefficients: (6 dropped because of singularities)
Estimate Std. Error t-value Pr(>|t|)
Trade_Chn_GDP -9.0284949 3.5674710 -2.5308 0.0646108 .
Trade_US_GDP 49.3859910 12.4808218 3.9570 0.0167203 *
Mil_China 0.0410458 0.0061361 6.6893 0.0025975 **
Mil_US0.1 0.3633651 0.3361052 1.0811 0.3404761
Mil_US0.1245 0.1343460 0.2654524 0.5061 0.6394041
Mil_US0.1452 0.5223941 0.2572227 2.0309 0.1120976
Mil_US0.15 0.7702547 0.1590245 4.8436 0.0083780 **
Mil_US0.2 -0.1590214 0.2829054 -0.5621 0.6040524
Mil_US0.2145 0.4225914 0.2335391 1.8095 0.1446296
Mil_US0.28 -0.0184684 0.2605845 -0.0709 0.9469007
Mil_US0.3 -0.1955169 0.3007200 -0.6502 0.5510402
Mil_US0.3214 -0.3582284 0.2421978 -1.4791 0.2132079
Mil_US0.35 0.1456503 0.2575910 0.5654 0.6019884
Mil_US0.42 -0.1028767 0.4173940 -0.2465 0.8174475
Mil_US10.2 -0.6257998 0.1869803 -3.3469 0.0286519 *
Mil_US102.12 -0.7389655 0.1831352 -4.0351 0.0156676 *
Mil_US102.5 -0.2628866 0.1792026 -1.4670 0.2162775
Mil_US105.45 -0.8838152 0.2313703 -3.8199 0.0187796 *
Mil_US108.12 -1.2636213 0.2459818 -5.1371 0.0068052 **
Mil_US11.45 -0.9917057 0.2672980 -3.7101 0.0206520 *
Mil_US110.2145 -0.8092059 0.3389560 -2.3873 0.0753847 .
Mil_US110.45 -1.1521668 0.2836680 -4.0617 0.0153276 *
Mil_US112.12 -0.6560746 0.2841673 -2.3088 0.0821439 .
Mil_US112.1452 -0.8874536 0.5436996 -1.6322 0.1779633
Mil_US115.4125 -1.3081285 0.4156094 -3.1475 0.0345973 *
Mil_US12.5112 -1.0846950 0.3331622 -3.2558 0.0312056 *
Mil_US120.15 1.9494307 0.3803047 5.1260 0.0068577 **
Mil_US120.5214 -0.5222347 0.4894904 -1.0669 0.3461329
Mil_US125.45 1.4025591 0.3681493 3.8098 0.0189441 *
Mil_US128.05 -0.6585287 0.5237435 -1.2573 0.2770325
Mil_US132.12 4.8290249 0.7167187 6.7377 0.0025286 **
Mil_US138.45 3.8219012 0.6999908 5.4599 0.0054705 **
Mil_US142.12 4.4277434 0.7859252 5.6338 0.0048845 **
Mil_US144.55 0.6882816 0.2759889 2.4939 0.0672053 .
Mil_US145.2412 1.4426232 0.3341803 4.3169 0.0124779 *
Mil_US148.4215 0.2779942 0.1831681 1.5177 0.2036960
Mil_US15.2452 -1.7939747 0.6493188 -2.7629 0.0507009 .
Mil_US152.1452 1.1249428 0.2892687 3.8889 0.0177068 *
Mil_US155.2 -2.3162524 0.2373424 -9.7591 0.0006176 ***
Mil_US160.2145 0.1680890 0.1779444 0.9446 0.3983422
Mil_US162.45 -2.5342309 0.2529421 -10.0190 0.0005579 ***
Mil_US168.45 -1.3159718 0.3073924 -4.2811 0.0128366 *
Mil_US170.8125 -2.2738274 0.3499096 -6.4983 0.0028928 **
Mil_US172.3 -1.2298963 0.2660813 -4.6223 0.0098644 **
Mil_US175.8 -1.4327347 0.3208578 -4.4653 0.0111146 *
Mil_US18.0125 -0.6187889 0.6073652 -1.0188 0.3659017
Mil_US180.12 -1.8262461 0.3186763 -5.7307 0.0045912 **
Mil_US195.521 0.8055906 0.2408536 3.3447 0.0287091 *
Mil_US20.1452 0.2744390 0.4969161 0.5523 0.6101598
Mil_US210.3341 -0.0416366 0.1783617 -0.2334 0.8268804
Mil_US22.45 0.1936566 0.5567873 0.3478 0.7455140
Mil_US240.2144 -1.0823252 0.2324999 -4.6552 0.0096243 **
Mil_US40.85 1.3761327 0.4269895 3.2229 0.0321926 *
Mil_US42.15 1.3486089 0.4289399 3.1441 0.0347121 *
Mil_US43.12 -0.4553799 0.2482606 -1.8343 0.1405276
Mil_US44.55 -0.2291647 0.2632385 -0.8706 0.4331228
Mil_US44.85 -0.5982856 0.2181709 -2.7423 0.0517839 .
Mil_US45.12 1.6619369 0.3721788 4.4654 0.0111137 *
Mil_US45.3125 -0.4334144 0.2419015 -1.7917 0.1476585
Mil_US48.7125 -0.4809525 0.1930415 -2.4914 0.0673800 .
Mil_US5.2 0.2849629 0.4923516 0.5788 0.5937653
Mil_US50.1452 -0.9064731 0.1871009 -4.8448 0.0083707 **
Mil_US52.4145 -0.2445888 0.1877570 -1.3027 0.2626176
Mil_US6.8 -0.1922175 0.3734064 -0.5148 0.6338535
Mil_US78.45 -1.6819336 0.2227979 -7.5491 0.0016496 **
Mil_US8.42 0.6691058 0.2516700 2.6587 0.0564690 .
Mil_US80.12 -1.5343597 0.2306790 -6.6515 0.0026528 **
Mil_US82.15 -0.2019613 0.2183042 -0.9251 0.4072557
Mil_US84.45 -0.7639822 0.2132468 -3.5826 0.0231135 *
Mil_US85.45 -0.7869678 0.2179931 -3.6101 0.0225554 *
Mil_US85.6452 -0.4424251 0.2178843 -2.0306 0.1121424
Mil_US86.12 -0.8500996 0.1938572 -4.3852 0.0118270 *
Mil_US88.4215 -0.9133360 0.2054882 -4.4447 0.0112927 *
Mil_US92.1124 -1.3818015 0.1894864 -7.2924 0.0018798 **
Mil_USn.a. 0.4868301 0.3000572 1.6225 0.1800255
Democracy 2.0123906 0.3626135 5.5497 0.0051578 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 32.301
Residual Sum of Squares: 0.061721
R-Squared: 0.99809
Adj. R-Squared: 0.95844
F-statistic: 27.4918 on 76 and 4 DF, p-value: 0.0025812
# 1. 安裝並載入格式化套件 if (! require (stargazer)) install.packages ("stargazer" )
Loading required package: 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
library (stargazer)# 2. 建立模型 (這裡以固定效應 FE 為例) # 確保變數名稱與你的 csv 一致 <- plm (Align ~ Trade_Chn_GDP + Trade_US_GDP + + Mil_US + + SCS, data = p_df, model = "within" )# 3. 輸出成像截圖中那樣的文字表格 stargazer (model_fe, type = "text" , title = "Regression Results: ASEAN Alignment Drivers" ,dep.var.labels = "Alignment Score (US vs China)" ,covariate.labels = c ("Trade Dependency (China)" , "Trade Dependency (US)" , "Military Aid (China)" , "Military Aid (US)" , "Democracy Index" , "SCS Dispute" ),out = "regression_results.txt" ) # 這會同時存成一個文字檔
Regression Results: ASEAN Alignment Drivers
======================================================
Dependent variable:
-----------------------------
Alignment Score (US vs China)
------------------------------------------------------
Trade Dependency (China) -9.028*
(3.567)
Trade Dependency (US) 49.386**
(12.481)
Military Aid (China) 0.041***
(0.006)
Military Aid (US) 0.363
(0.336)
Democracy Index 0.134
(0.265)
SCS Dispute 0.522
(0.257)
Mil_US0.15 0.770***
(0.159)
Mil_US0.2 -0.159
(0.283)
Mil_US0.2145 0.423
(0.234)
Mil_US0.28 -0.018
(0.261)
Mil_US0.3 -0.196
(0.301)
Mil_US0.3214 -0.358
(0.242)
Mil_US0.35 0.146
(0.258)
Mil_US0.42 -0.103
(0.417)
Mil_US10.2 -0.626**
(0.187)
Mil_US102.12 -0.739**
(0.183)
Mil_US102.5 -0.263
(0.179)
Mil_US105.45 -0.884**
(0.231)
Mil_US108.12 -1.264***
(0.246)
Mil_US11.45 -0.992**
(0.267)
Mil_US110.2145 -0.809*
(0.339)
Mil_US110.45 -1.152**
(0.284)
Mil_US112.12 -0.656*
(0.284)
Mil_US112.1452 -0.887
(0.544)
Mil_US115.4125 -1.308**
(0.416)
Mil_US12.5112 -1.085**
(0.333)
Mil_US120.15 1.949***
(0.380)
Mil_US120.5214 -0.522
(0.489)
Mil_US125.45 1.403**
(0.368)
Mil_US128.05 -0.659
(0.524)
Mil_US132.12 4.829***
(0.717)
Mil_US138.45 3.822***
(0.700)
Mil_US142.12 4.428***
(0.786)
Mil_US144.55 0.688*
(0.276)
Mil_US145.2412 1.443**
(0.334)
Mil_US148.4215 0.278
(0.183)
Mil_US15.2452 -1.794*
(0.649)
Mil_US152.1452 1.125**
(0.289)
Mil_US155.2 -2.316***
(0.237)
Mil_US160.2145 0.168
(0.178)
Mil_US162.45 -2.534***
(0.253)
Mil_US168.45 -1.316**
(0.307)
Mil_US170.8125 -2.274***
(0.350)
Mil_US172.3 -1.230***
(0.266)
Mil_US175.8 -1.433**
(0.321)
Mil_US18.0125 -0.619
(0.607)
Mil_US180.12 -1.826***
(0.319)
Mil_US195.521 0.806**
(0.241)
Mil_US20.1452 0.274
(0.497)
Mil_US210.3341 -0.042
(0.178)
Mil_US22.45 0.194
(0.557)
Mil_US240.2144 -1.082***
(0.232)
Mil_US40.85 1.376**
(0.427)
Mil_US42.15 1.349**
(0.429)
Mil_US43.12 -0.455
(0.248)
Mil_US44.55 -0.229
(0.263)
Mil_US44.85 -0.598*
(0.218)
Mil_US45.12 1.662**
(0.372)
Mil_US45.3125 -0.433
(0.242)
Mil_US48.7125 -0.481*
(0.193)
Mil_US5.2 0.285
(0.492)
Mil_US50.1452 -0.906***
(0.187)
Mil_US52.4145 -0.245
(0.188)
Mil_US6.8 -0.192
(0.373)
Mil_US78.45 -1.682***
(0.223)
Mil_US8.42 0.669*
(0.252)
Mil_US80.12 -1.534***
(0.231)
Mil_US82.15 -0.202
(0.218)
Mil_US84.45 -0.764**
(0.213)
Mil_US85.45 -0.787**
(0.218)
Mil_US85.6452 -0.442
(0.218)
Mil_US86.12 -0.850**
(0.194)
Mil_US88.4215 -0.913**
(0.205)
Mil_US92.1124 -1.382***
(0.189)
Mil_USn.a. 0.487
(0.300)
Democracy 2.012***
(0.363)
------------------------------------------------------
Observations 88
R2 0.998
Adjusted R2 0.958
F Statistic 27.492*** (df = 76; 4)
======================================================
Note: *p<0.1; **p<0.05; ***p<0.01
# 1. 載入套件 library (plm)library (stargazer)# 2. 建立三個不同層次的模型 # 模型 1: 經濟與軍事基礎模型 <- plm (Align ~ Trade_Chn_GDP + Trade_US_GDP + Mil_China + Mil_US, data = p_df, model = "within" )# 模型 2: 加入民主程度 <- plm (Align ~ Trade_Chn_GDP + Trade_US_GDP + Mil_China + Mil_US + Democracy, data = p_df, model = "within" )# 模型 3: 全模型 (包含南海爭議 SCS) <- plm (Align ~ Trade_Chn_GDP + Trade_US_GDP + Mil_China + Mil_US + Democracy + SCS, data = p_df, model = "within" )# 3. 輸出成像截圖那樣的專業表格 stargazer (m1, m2, m3, type = "text" , title = "Table 1: Determinants of ASEAN UN Voting Alignment (2014-2024)" ,dep.var.labels = "Alignment Score (Pro-US)" ,covariate.labels = c ("Trade w/ China (GDP%)" , "Trade w/ US (GDP%)" , "Military Aid (China)" , "Military Aid (US)" , "Democracy Index" , "SCS Dispute (Yes=1)" ),keep.stat = c ("n" , "rsq" , "adj.rsq" ), # 保留樣本數與 R2 notes = "Standard errors in parentheses: *p<0.1; **p<0.05; ***p<0.01" ,column.labels = c ("Economic/Mil" , "Institutional" , "Full Model" ))
Table 1: Determinants of ASEAN UN Voting Alignment (2014-2024)
===================================================================================
Dependent variable:
-------------------------------------------------------------
Alignment Score (Pro-US)
Economic/Mil Institutional Full Model
(1) (2) (3)
-----------------------------------------------------------------------------------
Trade w/ China (GDP%) -17.502* -9.028* -9.028*
(8.506) (3.567) (3.567)
Trade w/ US (GDP%) 62.539 49.386** 49.386**
(32.327) (12.481) (12.481)
Military Aid (China) 0.027 0.041*** 0.041***
(0.015) (0.006) (0.006)
Military Aid (US) -0.836 0.363 0.363
(0.679) (0.336) (0.336)
Democracy Index -0.603 0.134 0.134
(0.606) (0.265) (0.265)
SCS Dispute (Yes=1) -0.150 0.522 0.522
(0.598) (0.257) (0.257)
Mil_US0.15 0.763 0.770*** 0.770***
(0.420) (0.159) (0.159)
Mil_US0.2 -0.593 -0.159 -0.159
(0.717) (0.283) (0.283)
Mil_US0.2145 -0.418 0.423 0.423
(0.469) (0.234) (0.234)
Mil_US0.28 1.043* -0.018 -0.018
(0.467) (0.261) (0.261)
Mil_US0.3 -0.782 -0.196 -0.196
(0.743) (0.301) (0.301)
Mil_US0.3214 -1.028 -0.358 -0.358
(0.554) (0.242) (0.242)
Mil_US0.35 0.966 0.146 0.146
(0.557) (0.258) (0.258)
Mil_US0.42 1.664* -0.103 -0.103
(0.712) (0.417) (0.417)
Mil_US10.2 -0.560 -0.626** -0.626**
(0.492) (0.187) (0.187)
Mil_US102.12 -0.502 -0.739** -0.739**
(0.470) (0.183) (0.183)
Mil_US102.5 -0.232 -0.263 -0.263
(0.473) (0.179) (0.179)
Mil_US105.45 -0.156 -0.884** -0.884**
(0.503) (0.231) (0.231)
Mil_US108.12 -0.684 -1.264*** -1.264***
(0.587) (0.246) (0.246)
Mil_US11.45 -0.973 -0.992** -0.992**
(0.705) (0.267) (0.267)
Mil_US110.2145 -0.860 -0.809* -0.809*
(0.894) (0.339) (0.339)
Mil_US110.45 -0.448 -1.152** -1.152**
(0.669) (0.284) (0.284)
Mil_US112.12 -0.688 -0.656* -0.656*
(0.750) (0.284) (0.284)
Mil_US112.1452 -1.016 -0.887 -0.887
(1.433) (0.544) (0.544)
Mil_US115.4125 -1.048 -1.308** -1.308**
(1.089) (0.416) (0.416)
Mil_US12.5112 -1.394 -1.085** -1.085**
(0.867) (0.333) (0.333)
Mil_US120.15 1.280 1.949*** 1.949***
(0.952) (0.380) (0.380)
Mil_US120.5214 -0.739 -0.522 -0.522
(1.287) (0.489) (0.489)
Mil_US125.45 0.772 1.403** 1.403**
(0.924) (0.368) (0.368)
Mil_US128.05 -0.856 -0.659 -0.659
(1.379) (0.524) (0.524)
Mil_US132.12 1.259 4.829*** 4.829***
(0.834) (0.717) (0.717)
Mil_US138.45 0.271 3.822*** 3.822***
(0.749) (0.700) (0.700)
Mil_US142.12 0.330 4.428*** 4.428***
(0.710) (0.786) (0.786)
Mil_US144.55 -0.002 0.688* 0.688*
(0.650) (0.276) (0.276)
Mil_US145.2412 0.234 1.443** 1.443**
(0.669) (0.334) (0.334)
Mil_US148.4215 0.210 0.278 0.278
(0.482) (0.183) (0.183)
Mil_US15.2452 -1.929 -1.794* -1.794*
(1.712) (0.649) (0.649)
Mil_US152.1452 -0.041 1.125** 1.125**
(0.525) (0.289) (0.289)
Mil_US155.2 -1.866** -2.316*** -2.316***
(0.588) (0.237) (0.237)
Mil_US160.2145 0.110 0.168 0.168
(0.469) (0.178) (0.178)
Mil_US162.45 -1.807** -2.534*** -2.534***
(0.571) (0.253) (0.253)
Mil_US168.45 -0.466 -1.316** -1.316**
(0.703) (0.307) (0.307)
Mil_US170.8125 -0.855 -2.274*** -2.274***
(0.631) (0.350) (0.350)
Mil_US172.3 -0.535 -1.230*** -1.230***
(0.619) (0.266) (0.266)
Mil_US175.8 -0.601 -1.433** -1.433**
(0.748) (0.321) (0.321)
Mil_US18.0125 -1.358 -0.619 -0.619
(1.563) (0.607) (0.607)
Mil_US180.12 -0.709 -1.826*** -1.826***
(0.652) (0.319) (0.319)
Mil_US195.521 -0.090 0.806** 0.806**
(0.472) (0.241) (0.241)
Mil_US20.1452 -0.738 0.274 0.274
(1.219) (0.497) (0.497)
Mil_US210.3341 -0.028 -0.042 -0.042
(0.471) (0.178) (0.178)
Mil_US22.45 -0.791 0.194 0.194
(1.392) (0.557) (0.557)
Mil_US240.2144 -0.418 -1.082*** -1.082***
(0.526) (0.232) (0.232)
Mil_US40.85 -0.092 1.376** 1.376**
(0.884) (0.427) (0.427)
Mil_US42.15 -0.040 1.349** 1.349**
(0.919) (0.429) (0.429)
Mil_US43.12 -0.416 -0.455 -0.455
(0.655) (0.248) (0.248)
Mil_US44.55 -0.238 -0.229 -0.229
(0.694) (0.263) (0.263)
Mil_US44.85 -0.586 -0.598* -0.598*
(0.576) (0.218) (0.218)
Mil_US45.12 0.427 1.662** 1.662**
(0.787) (0.372) (0.372)
Mil_US45.3125 -0.403 -0.433 -0.433
(0.638) (0.242) (0.242)
Mil_US48.7125 -0.128 -0.481* -0.481*
(0.481) (0.193) (0.193)
Mil_US5.2 0.962 0.285 0.285
(1.258) (0.492) (0.492)
Mil_US50.1452 -0.601 -0.906*** -0.906***
(0.472) (0.187) (0.187)
Mil_US52.4145 0.045 -0.245 -0.245
(0.476) (0.188) (0.188)
Mil_US6.8 0.686 -0.192 -0.192
(0.892) (0.373) (0.373)
Mil_US78.45 -1.306* -1.682*** -1.682***
(0.560) (0.223) (0.223)
Mil_US8.42 1.270* 0.669* 0.669*
(0.599) (0.252) (0.252)
Mil_US80.12 -0.927 -1.534*** -1.534***
(0.536) (0.231) (0.231)
Mil_US82.15 -0.792 -0.202 -0.202
(0.503) (0.218) (0.218)
Mil_US84.45 -1.323** -0.764** -0.764**
(0.496) (0.213) (0.213)
Mil_US85.45 -0.256 -0.787** -0.787**
(0.517) (0.218) (0.218)
Mil_US85.6452 -1.077* -0.442 -0.442
(0.489) (0.218) (0.218)
Mil_US86.12 -1.184* -0.850** -0.850**
(0.486) (0.194) (0.194)
Mil_US88.4215 -0.640 -0.913** -0.913**
(0.526) (0.205) (0.205)
Mil_US92.1124 -1.071* -1.382*** -1.382***
(0.478) (0.189) (0.189)
Mil_USn.a. -0.348 0.487 0.487
(0.685) (0.300) (0.300)
Democracy 2.012*** 2.012***
(0.363) (0.363)
-----------------------------------------------------------------------------------
Observations 88 88 88
R2 0.983 0.998 0.998
Adjusted R2 0.711 0.958 0.958
===================================================================================
Note: *p<0.1; **p<0.05; ***p<0.01
Standard errors in parentheses: *p<0.1; **p<0.05; ***p<0.01
library (ggplot2)# 繪製四分位圖 ggplot (df, aes (x = factor (Year), y = Align, fill = factor (Year))) + geom_boxplot (alpha = 0.7 ) + geom_jitter (width = 0.2 , alpha = 0.5 ) + theme_minimal () + labs (title = "ASEAN Alignment Score Distribution (2014-2024)" ,x = "Year" , y = "Align" ) + theme (legend.position = "none" )
# 1. 載入套件 library (tidyverse)library (plm)library (stargazer)# 2. 讀取並「深度清洗」數據 <- read.csv ("aseananalysis.csv" , stringsAsFactors = FALSE )# --- 關鍵步驟：把所有 "n.a." 或文字轉為數字，並處理缺失值 --- # 我們將核心變數強制轉換為數值，非數字的會變成 NA <- df %>% mutate (across (c (Align, Mil_US, Mil_China, Trade_Chn_GDP, Trade_US_GDP, Democracy), ~ as.numeric (gsub ("[^0-9.-]" , "" , .))))
Warning: There was 1 warning in `mutate()`.
ℹ In argument: `across(...)`.
Caused by warning:
! NAs introduced by coercion
# 剔除包含 NA 的列，確保回歸模型不會因為 NA 而出錯 <- na.omit (df)# 3. 建立面板數據格式 <- pdata.frame (df_clean, index = c ("Country" , "Year" ))# 4. 重新跑模型 (確保所有變數都是數字) <- plm (Align ~ Mil_US + Mil_China, data = p_df, model = "within" )<- plm (Align ~ Mil_US + Mil_China + Trade_US_GDP + Trade_Chn_GDP, data = p_df, model = "within" )<- plm (Align ~ Mil_US + Mil_China + Trade_US_GDP + Trade_Chn_GDP + Democracy + SCS, data = p_df, model = "within" )# 5. 輸出表格 (先不加標籤，測試是否成功) # 如果這次成功了，再慢慢把 labels 加回去 stargazer (res1, res2, res3, type = "text" , title = "Regression Analysis of ASEAN Alignment" ,column.labels = c ("Model 1" , "Model 2" , "Model 3" ))
Regression Analysis of ASEAN Alignment
==================================================================================
Dependent variable:
--------------------------------------------------------------------
Align
Model 1 Model 2 Model 3
(1) (2) (3)
----------------------------------------------------------------------------------
Mil_US 0.021*** 0.017*** 0.017***
(0.004) (0.004) (0.004)
Mil_China 0.023*** 0.018*** 0.019***
(0.005) (0.005) (0.005)
Trade_US_GDP 32.511*** 32.301***
(5.263) (5.315)
Trade_Chn_GDP -16.359*** -16.028***
(3.281) (3.383)
Democracy 0.055
(0.123)
----------------------------------------------------------------------------------
Observations 82 82 82
R2 0.417 0.624 0.625
Adjusted R2 0.344 0.564 0.559
F Statistic 25.734*** (df = 2; 72) 28.995*** (df = 4; 70) 22.969*** (df = 5; 69)
==================================================================================
Note: *p<0.1; **p<0.05; ***p<0.01
<- plm (Align ~ Mil_US + Mil_China + Trade_US_GDP + Trade_Chn_GDP + Democracy + SCS, data = p_df, model = "pooling" )summary (res_ols)
Pooling Model
Call:
plm(formula = Align ~ Mil_US + Mil_China + Trade_US_GDP + Trade_Chn_GDP +
Democracy + SCS, data = p_df, model = "pooling")
Unbalanced Panel: n = 8, T = 5-11, N = 82
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-1.14119 -0.39514 -0.16007 0.47550 1.19221
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
(Intercept) -4.4416073 0.8493106 -5.2297 1.485e-06 ***
Mil_US 0.0066043 0.0016120 4.0969 0.0001048 ***
Mil_China 0.0054695 0.0024596 2.2238 0.0291745 *
Trade_US_GDP 4.5735251 2.6539712 1.7233 0.0889592 .
Trade_Chn_GDP 1.4378689 2.5605944 0.5615 0.5761056
Democracy 0.0290944 0.0805729 0.3611 0.7190441
SCS 0.0814135 0.1724115 0.4722 0.6381520
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 35.338
Residual Sum of Squares: 25.087
R-Squared: 0.2901
Adj. R-Squared: 0.2333
F-statistic: 5.10799 on 6 and 75 DF, p-value: 0.00019103
檢驗貢獻性
Loading required package: carData
The following object is masked from 'package:dplyr':
recode
The following object is masked from 'package:purrr':
some
# 注意：plm 模型不能直接跑 vif，要轉成普通 lm 跑一下看相關性 <- lm (Align ~ Mil_US + Mil_China + Trade_US_GDP + Trade_Chn_GDP + Democracy + SCS, data = df_clean)vif (vif_check)
Mil_US Mil_China Trade_US_GDP Trade_Chn_GDP Democracy
2.849601 2.794066 3.749078 10.458091 4.830572
SCS
1.812050
# 使用 lag 函數 <- plm (Align ~ lag (Mil_US) + lag (Mil_China) + lag (Trade_Chn_GDP) + SCS, data = p_df, model = "within" )summary (res_lag)
Oneway (individual) effect Within Model
Call:
plm(formula = Align ~ lag(Mil_US) + lag(Mil_China) + lag(Trade_Chn_GDP) +
SCS, data = p_df, model = "within")
Unbalanced Panel: n = 8, T = 4-10, N = 74
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-0.852143 -0.395867 -0.038797 0.346545 1.076137
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
lag(Mil_US) 0.0279098 0.0059420 4.6970 1.475e-05 ***
lag(Mil_China) 0.0232148 0.0074638 3.1103 0.002806 **
lag(Trade_Chn_GDP) 2.2080885 2.8343964 0.7790 0.438877
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 28.444
Residual Sum of Squares: 15.567
R-Squared: 0.45271
Adj. R-Squared: 0.36584
F-statistic: 17.3708 on 3 and 63 DF, p-value: 2.4911e-08
library (plm)library (stargazer)# 1. 數據清洗與標準化 (Standardization) # 將數值變數標準化，這能讓係數更易於比較，並優化模型擬合度 <- df_clean %>% mutate (across (c (Mil_US, Mil_China, Trade_US_GDP, Trade_Chn_GDP, Democracy), ~ as.numeric (scale (.))))<- pdata.frame (df_final, index = c ("Country" , "Year" ))# 2. 建立三個模型 # Model 1: 安全與軍事 (Realism perspective) <- plm (Align ~ Mil_US + Mil_China, data = p_df_final, model = "within" )# Model 2: 經濟與制度 (Liberalism perspective) <- plm (Align ~ Trade_US_GDP + Trade_Chn_GDP + Democracy, data = p_df_final, model = "within" )# Model 3: 綜合地緣模型 (包含 SCS) <- plm (Align ~ Mil_US + Mil_China + Trade_US_GDP + Trade_Chn_GDP + Democracy + SCS, data = p_df_final, model = "within" )# 3. 輸出正式表格 stargazer (m1, m2, m3, type = "text" , title = "Determinants of ASEAN UN Voting Alignment" ,dep.var.labels = "Alignment Score" ,covariate.labels = c ("Military Aid (US)" , "Military Aid (China)" , "Trade (US % GDP)" , "Trade (China % GDP)" , "Democracy Index" , "SCS Dispute (Yes=1)" ),column.labels = c ("Security" , "Econ & Inst" , "Full Model" ),digits = 3 ,star.cutoffs = c (0.05 , 0.01 , 0.001 ), # 使用更嚴謹的顯著性標準 notes = "Normalized coefficients. Standard errors in parentheses." )
Determinants of ASEAN UN Voting Alignment
=========================================================================================
Dependent variable:
--------------------------------------------------------------------
Alignment Score
Security Econ Inst
(1) (2) (3)
-----------------------------------------------------------------------------------------
Military Aid (US) 1.424*** 1.158***
(0.285) (0.250)
Military Aid (China) 1.005*** 0.815***
(0.237) (0.223)
Trade (US % GDP) 1.925*** 1.514***
(0.289) (0.249)
Trade (China % GDP) -1.170*** -1.301***
(0.331) (0.275)
Democracy Index -0.046 0.096
(0.252) (0.216)
-----------------------------------------------------------------------------------------
Observations 82 82 82
R2 0.417 0.436 0.625
Adjusted R2 0.344 0.356 0.559
F Statistic 25.734*** (df = 2; 72) 18.284*** (df = 3; 71) 22.969*** (df = 5; 69)
=========================================================================================
Note: *p<0.05; **p<0.01; ***p<0.001
Normalized coefficients. Standard errors in parentheses.
# 看看 res3 實際產出的變數名稱和數量 names (coef (res3))
[1] "Mil_US" "Mil_China" "Trade_US_GDP" "Trade_Chn_GDP"
[5] "Democracy"
# 1. 強制轉換資料型態，確保沒有文字干擾 <- df %>% mutate (across (c (Align, Mil_US, Mil_China, Trade_US_GDP, Trade_Chn_GDP, Democracy), ~ as.numeric (as.character (.)))) %>% na.omit ()<- pdata.frame (df_clean, index = c ("Country" , "Year" ))# 2. 重新跑模型 <- plm (Align ~ Mil_US + Mil_China, data = p_df, model = "within" )<- plm (Align ~ Mil_US + Mil_China + Trade_US_GDP + Trade_Chn_GDP + Democracy, data = p_df, model = "within" )<- plm (Align ~ Mil_US + Mil_China + Trade_US_GDP + Trade_Chn_GDP + Democracy + SCS, data = p_df, model = "within" )
# 載入必要的套件 library (ggplot2)library (plm)library (sjPlot)
Attaching package: 'sjPlot'
The following object is masked from 'package:ggplot2':
set_theme
library (readr) # 1. 正確讀取資料 (請確認引號內的檔名跟你的資料夾裡的一模一樣) <- read_csv ("aseananalysis.csv" )
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (2): Country, Mil_US
dbl (7): Year, Align, Mil_China, Trade_Chn_GDP, Trade_US_GDP, Democracy, SCS
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
# 2. 數據清洗與轉換 (這步很重要，否則畫圖會出錯) $ Align <- as.numeric (as.character (asean_clean$ Align))$ Trade_Chn_GDP <- as.numeric (as.character (asean_clean$ Trade_Chn_GDP))$ Trade_US_GDP <- as.numeric (as.character (asean_clean$ Trade_US_GDP))$ Mil_China <- as.numeric (as.character (asean_clean$ Mil_China))$ Mil_US <- as.numeric (as.character (asean_clean$ Mil_US))
Warning: NAs introduced by coercion
$ Democracy <- as.numeric (as.character (asean_clean$ Democracy))# 3. 定義面板與模型 <- pdata.frame (asean_clean, index = c ("Country" , "Year" ))<- plm (Align ~ Trade_Chn_GDP + Trade_US_GDP + + Mil_US + + SCS, data = p_df_full, model = "within" )# 4. 執行森林圖 plot_model (model_full, show.values = TRUE , value.offset = .3 ,title = "Determinants of ASEAN Strategic Alignment" ,axis.labels = c ("Democracy Index" , "Mil (US)" , "Mil (China)" , "Trade (US)" , "Trade (China)" )) + theme_minimal ()
Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
ℹ Please use `linewidth` instead.
ℹ The deprecated feature was likely used in the sjPlot package.
Please report the issue at <https://github.com/strengejacke/sjPlot/issues>.
# 看看對美貿易與對齊度的關係 ggplot (asean_clean, aes (x = Trade_US_GDP, y = Align)) + geom_point (aes (color = Country), size = 3 ) + geom_smooth (method = "lm" , color = "black" , linetype = "dashed" ) + # 加入回歸線 theme_minimal () + labs (title = "Relationship between US Trade Dependence and Alignment" ,x = "Trade dependence (US) % of GDP" ,y = "Strategic Alignment Score" )
`geom_smooth()` using formula = 'y ~ x'
# 1. 確保軍事數據是數值型態 $ Mil_US <- as.numeric (as.character (asean_clean$ Mil_US))# 2. 繪製散佈圖與回歸線 library (ggplot2)ggplot (asean_clean, aes (x = Mil_US, y = Align)) + geom_point (aes (color = Country), size = 3 , alpha = 0.7 ) + # alpha 增加透明度，避免點重疊 geom_smooth (method = "lm" , color = "darkblue" , linetype = "solid" ) + # 改為深藍實線增加辨識度 theme_minimal () + labs (title = "Relationship between US Military Cooperation and Alignment" ,subtitle = "ASEAN Strategic Alignment (2014-2023)" ,x = "US Military Expenditure/Cooperation (Value)" ,y = "Strategic Alignment Score" ) + theme (legend.position = "right" )
`geom_smooth()` using formula = 'y ~ x'
Warning: Removed 6 rows containing non-finite outside the scale range
(`stat_smooth()`).
Warning: Removed 6 rows containing missing values or values outside the scale range
(`geom_point()`).
# 1. 確保數據格式正確 $ Trade_Chn_GDP <- as.numeric (as.character (asean_clean$ Trade_Chn_GDP))# 2. 繪製散佈圖（對中貿易 vs 對齊度） library (ggplot2)ggplot (asean_clean, aes (x = Trade_Chn_GDP, y = Align)) + geom_point (aes (color = Country), size = 3 , alpha = 0.7 ) + # 使用紅色系代表中國因素，並加入回歸線 geom_smooth (method = "lm" , color = "darkred" , linetype = "dashed" ) + theme_minimal () + labs (title = "Relationship between China Trade Dependence and Alignment" ,subtitle = "Negative correlation: Higher trade dependence leads to more pro-China alignment" ,x = "Trade dependence (China) % of GDP" ,y = "Strategic Alignment Score" ) + theme (legend.position = "right" )
`geom_smooth()` using formula = 'y ~ x'
ggplot (asean_clean, aes (x = Trade_US_GDP, y = Trade_Chn_GDP)) + geom_path (aes (group = Country, color = Country), arrow = arrow (length = unit (0.2 ,"cm" ))) + # 畫出路徑 geom_point (aes (size = abs (Align))) + theme_minimal () + labs (title = "Trade Entrapment: Dual Dependence Trajectory" ,x = "Dependence on US" , y = "Dependence on China" )
# 計算每年對齊分數的分散程度 <- aggregate (Align ~ Year, data = asean_clean, FUN = sd)ggplot (polarization, aes (x = Year, y = Align)) + geom_line (size = 1.2 , color = "darkred" ) + geom_point (size = 3 ) + theme_minimal () + labs (title = "Diplomatic Polarization in ASEAN" ,subtitle = "Increasing SD means countries are being pushed to opposite poles" ,y = "Standard Deviation of Alignment" , x = "Year" )
# 計算每年東協整體的平均立場與標準差 <- aggregate (Align ~ Year, data = asean_clean, FUN = function (x) c (mean = mean (x), sd = sd (x)))<- do.call (data.frame, asean_consensus)# 畫出標準差隨時間的變化 library (ggplot2)ggplot (asean_consensus, aes (x = Year, y = Align.sd)) + geom_line (size = 1.2 , color = "darkgreen" ) + geom_point (size = 3 ) + theme_minimal () + labs (title = "Is ASEAN Consensus Collapsing?" ,subtitle = "Higher Standard Deviation means increasing internal polarization" ,x = "Year" , y = "Standard Deviation of Alignment Score" )
library (ggplot2)ggplot (asean_clean, aes (x = as.factor (Year), y = Align)) + geom_boxplot (fill = "lightgray" , outlier.shape = NA ) + # 畫出整體的範圍 geom_jitter (aes (color = Country), width = 0.2 , size = 2 ) + # 點出每個國家的位置 theme_minimal () + labs (title = "ASEAN Internal Alignment Consensus (2014-2023)" ,subtitle = "Wider vertical spread indicates declining consensus" ,x = "Year" , y = "Strategic Alignment Score" )
# 假設你已經根據 Democracy 指數將國家分類 (例如：High vs Low) # 如果還沒分類，我們可以用中位數簡單切分： $ Regime_Type <- ifelse (asean_clean$ Democracy > median (asean_clean$ Democracy, na.rm= T), "Higher Democracy" , "Lower Democracy" )ggplot (asean_clean, aes (x = Regime_Type, y = Align, fill = Regime_Type)) + geom_boxplot (alpha = 0.7 ) + geom_jitter (width = 0.1 ) + theme_minimal () + labs (title = "Impact of Regime Heterogeneity on Alignment" ,x = "Regime Type" , y = "Strategic Alignment Score" )
# 馬來西亞的數據 <- subset (asean_clean, Country == "Malaysia" )ggplot (malaysia_data, aes (x = Year)) + geom_line (aes (y = Align, color = "Strategic Alignment" ), size = 1.2 ) + geom_line (aes (y = Trade_Chn_GDP / 10 , color = "China Trade (scaled)" ), linetype = "dashed" ) + # 縮小比例方便對比 theme_minimal () + labs (title = "The Hedging Profile of Malaysia" ,subtitle = "Maintaining stability despite economic fluctuations" ,y = "Value" , x = "Year" )
# 越南的數據 <- subset (asean_clean, Country == "Vietnam" )ggplot (malaysia_data, aes (x = Year)) + geom_line (aes (y = Align, color = "Strategic Alignment" ), size = 1.2 ) + geom_line (aes (y = Trade_Chn_GDP / 10 , color = "China Trade (scaled)" ), linetype = "dashed" ) + # 縮小比例方便對比 theme_minimal () + labs (title = "The Hedging Profile of vietnam" ,subtitle = "Maintaining stability despite economic fluctuations" ,y = "Value" , x = "Year" )
# 篩選菲律賓數據 <- subset (asean_clean, Country == "Philippines" )ggplot (philippines_data, aes (x = Year)) + # 畫出對齊分數（實線） geom_line (aes (y = Align, color = "Strategic Alignment" ), size = 1.5 ) + geom_point (aes (y = Align, color = "Strategic Alignment" ), size = 3 ) + # 畫出對中貿易（標準化後） geom_line (aes (y = (Trade_Chn_GDP - mean (Trade_Chn_GDP))/ sd (Trade_Chn_GDP), color = "China Trade (Standardized)" ), linetype = "dotted" , size = 1 ) + # 加入政治事件標記 geom_vline (xintercept = 2016 , linetype = "dashed" , color = "darkred" , alpha = 0.5 ) + # 杜特蒂上任 geom_vline (xintercept = 2022 , linetype = "dashed" , color = "darkblue" , alpha = 0.5 ) + # 小馬可仕上任 annotate ("text" , x = 2018 , y = - 1 , label = "Duterte Era \n (Pro-China Pivot)" , size = 3 ) + annotate ("text" , x = 2023 , y = 1 , label = "Marcos Jr. Era \n (Back to US)" , size = 3 ) + theme_minimal () + scale_color_manual (values = c ("Strategic Alignment" = "#2980B9" , "China Trade (Standardized)" = "#C0392B" )) + labs (title = "The Philippine Pendulum: From Pivot to Realignment" ,subtitle = "Strategic alignment shifts driven by leadership changes" ,y = "Alignment Score" , x = "Year" )
library (ggplot2)# 1. 篩選具代表性的四個國家 <- c ("Malaysia" , "Vietnam" , "Philippines" , "Cambodia" )<- subset (asean_clean, Country %in% target_countries)# 2. 繪製對比圖 ggplot (comparison_df, aes (x = Year, y = Align, color = Country, group = Country)) + # 加入背景中立線 geom_hline (yintercept = 0 , linetype = "dashed" , color = "gray60" , alpha = 0.5 ) + # 畫出平滑趨勢線與數據點 geom_line (size = 1.2 , alpha = 0.8 ) + geom_point (size = 3 ) + # 使用專業色調 scale_color_manual (values = c ("Malaysia" = "#27ae60" , # 穩定的綠色（對沖/中道） "Vietnam" = "#2980b9" , # 深藍色（韌性/抗衡） "Philippines" = "#f39c12" , # 橘色（劇烈擺盪） "Cambodia" = "#c0392b" # 深紅色（高度依賴/下沉） + # 加上圖表資訊 labs (title = "Strategic Trajectories of Selected ASEAN States (2014-2023)" ,subtitle = "Contrasting Hedging, Balancing, and Bandwagoning Behaviors" ,x = "Year" , y = "Alignment Score (Positive: Pro-US | Negative: Pro-China)" ,color = "Country Case" + theme_minimal (base_size = 14 ) + theme (legend.position = "bottom" )
