Strategic Ambiguity

# =========================
# 0. 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

library(janitor)

Attaching package: 'janitor'

The following objects are masked from 'package:stats':

    chisq.test, fisher.test

library(ggplot2)
library(corrplot)

corrplot 0.95 loaded

library(broom)
library(plm)

Attaching package: 'plm'

The following objects are masked from 'package:dplyr':

    between, lag, lead

# =========================
# 1. 讀取資料
# =========================

df <- read_excel("~/Desktop/data/Strategic Ambiguity.xlsx") %>%
  clean_names()

names(df)

[1] "country"             "year"                "political_stability"
[4] "democracy"           "scs"                 "post2022"           
[7] "trade_china"         "sec_us"              "strategic_ambiguity"

# =========================
# 2. 戰略模糊折線圖
# =========================

ggplot(df, aes(x = year, y = strategic_ambiguity, 
               color = country, group = country)) +
  geom_line(linewidth = 1.1) +
  geom_point(size = 2) +
  labs(
    title = "Strategic Ambiguity Trends among ASEAN Countries",
    x = "Year",
    y = "Strategic Ambiguity",
    color = "Country"
  ) +
  theme_minimal(base_size = 13) +
  theme(
    plot.title = element_text(face = "bold", size = 15),
    legend.position = "right"
  )

ggplot(df, aes(x = year, y = strategic_ambiguity, 
               color = country, group = country)) +
  geom_line(linewidth = 1.1) +
  geom_point(size = 2) +
  geom_vline(xintercept = 2022, linetype = "dashed", color = "gray40") +
  annotate("text", x = 2022.1, y = max(df$strategic_ambiguity, na.rm = TRUE),
           label = "Post-2022", hjust = 0, size = 4) +
  labs(
    title = "Strategic Ambiguity before and after 2022",
    x = "Year",
    y = "Strategic Ambiguity",
    color = "Country"
  ) +
  theme_minimal(base_size = 13)

# =========================
# 3. 相關係數圖
# =========================

cor_data <- df %>%
  select(
    strategic_ambiguity,
    political_stability,
    democracy,
    scs,
    post2022,
    trade_china,
    sec_us
  ) %>%
  mutate(across(everything(), as.numeric))

cor_matrix <- cor(cor_data, use = "complete.obs")

corrplot(
  cor_matrix,
  method = "color",
  type = "upper",
  addCoef.col = "black",
  tl.col = "black",
  tl.srt = 45,
  number.cex = 0.8,
  title = "Correlation Matrix",
  mar = c(0, 0, 2, 0)
)

cor_long <- as.data.frame(cor_matrix) %>%
  rownames_to_column("var1") %>%
  pivot_longer(-var1, names_to = "var2", values_to = "correlation")

ggplot(cor_long, aes(x = var1, y = var2, fill = correlation)) +
  geom_tile(color = "white") +
  geom_text(aes(label = round(correlation, 2)), size = 4) +
  scale_fill_gradient2(
    low = "#2166AC",
    mid = "white",
    high = "#B2182B",
    midpoint = 0,
    limits = c(-1, 1)
  ) +
  labs(
    title = "Correlation Heatmap",
    x = "",
    y = "",
    fill = "Correlation"
  ) +
  theme_minimal(base_size = 13) +
  theme(
    axis.text.x = element_text(angle = 45, hjust = 1),
    plot.title = element_text(face = "bold", size = 15)
  )

# =========================
# 4. OLS 回歸模型
# =========================

model_ols <- lm(
  strategic_ambiguity ~ political_stability + democracy + scs + post2022,
  data = df
)

summary(model_ols)

Call:
lm(formula = strategic_ambiguity ~ political_stability + democracy + 
    scs + post2022, data = df)

Residuals:
    Min      1Q  Median      3Q     Max 
-42.770 -10.692   3.055  12.505  35.446 

Coefficients:
                    Estimate Std. Error t value Pr(>|t|)    
(Intercept)          34.8375    25.0269   1.392 0.172705    
political_stability   0.0058     0.3129   0.019 0.985314    
democracy             7.0189     1.7532   4.004 0.000309 ***
scs                 -14.8289     7.3357  -2.021 0.050929 .  
post2022             10.4480     6.5032   1.607 0.117129    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 20.1 on 35 degrees of freedom
Multiple R-squared:  0.3594,    Adjusted R-squared:  0.2862 
F-statistic: 4.909 on 4 and 35 DF,  p-value: 0.003005

# =========================
# 6. 政治穩定度與戰略模糊散佈圖
# =========================

ggplot(df, aes(x = political_stability, y = strategic_ambiguity)) +
  geom_point(aes(color = country), size = 3, alpha = 0.8) +
  geom_smooth(method = "lm", se = TRUE, color = "black") +
  labs(
    title = "Political Stability and Strategic Ambiguity",
    x = "Political Stability",
    y = "Strategic Ambiguity",
    color = "Country"
  ) +
  theme_minimal(base_size = 13) +
  theme(
    plot.title = element_text(face = "bold", size = 15)
  )

`geom_smooth()` using formula = 'y ~ x'

# =========================
# 7. Panel Data
# =========================

p_df <- pdata.frame(df, index = c("country", "year"))

model_fe <- plm(
  strategic_ambiguity ~ political_stability + democracy + post2022,
  data = p_df,
  model = "within",
  effect = "individual"
)

summary(model_fe)

Oneway (individual) effect Within Model

Call:
plm(formula = strategic_ambiguity ~ political_stability + democracy + 
    post2022, data = p_df, effect = "individual", model = "within")

Balanced Panel: n = 8, T = 5, N = 40

Residuals:
    Min.  1st Qu.   Median  3rd Qu.     Max. 
-23.5751  -9.4914  -1.0830   6.3004  36.1416 

Coefficients:
                    Estimate Std. Error t-value Pr(>|t|)  
political_stability -0.21692    1.68704 -0.1286  0.89858  
democracy            1.35218   12.55351  0.1077  0.91496  
post2022            10.64700    5.59222  1.9039  0.06689 .
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Total Sum of Squares:    7958.8
Residual Sum of Squares: 6935
R-Squared:      0.12863
Adj. R-Squared: -0.17184
F-statistic: 1.42698 on 3 and 29 DF, p-value: 0.25505

fe_result <- tidy(model_fe, conf.int = TRUE)

ggplot(fe_result, aes(x = estimate, y = reorder(term, estimate))) +
  geom_point(size = 3) +
  geom_errorbarh(aes(xmin = conf.low, xmax = conf.high), height = 0.2) +
  geom_vline(xintercept = 0, linetype = "dashed", color = "gray40") +
  labs(
    title = "Fixed Effects Regression Coefficients",
    x = "Coefficient Estimate",
    y = "Variables"
  ) +
  theme_minimal(base_size = 13)

Warning: `geom_errorbarh()` was deprecated in ggplot2 4.0.0.
ℹ Please use the `orientation` argument of `geom_errorbar()` instead.

`height` was translated to `width`.

# 1. 載入套件
library(readxl)
library(tidyverse)
library(plm)
library(stargazer)

Please cite as: 

 Hlavac, Marek (2022). stargazer: Well-Formatted Regression and Summary Statistics Tables.

 R package version 5.2.3. https://CRAN.R-project.org/package=stargazer 

# 2. 讀取資料
df <- read_excel("~/Desktop/data/Strategic Ambiguity.xlsx")

# 3. 整理欄位名稱
names(df) <- names(df) %>%
  str_replace_all("＿", "_") %>%
  str_replace_all(" ", "_") %>%
  str_to_lower()

# 4. 確認欄位名稱
names(df)

[1] "country"             "year"                "political_stability"
[4] "democracy"           "scs"                 "post2022"           
[7] "trade_china"         "sec_us"              "strategic_ambiguity"

# 5. 確保變數型態正確
df <- df %>%
  mutate(
    country = as.factor(country),
    year = as.numeric(year),
    strategic_ambiguity = as.numeric(strategic_ambiguity),
    political_stability = as.numeric(political_stability),
    democracy = as.numeric(democracy),
    scs = as.factor(scs),
    post2022 = as.factor(post2022)
  )

# 6. OLS 模型
ols_model <- lm(
  strategic_ambiguity ~ political_stability + democracy + scs + post2022,
  data = df
)

summary(ols_model)

Call:
lm(formula = strategic_ambiguity ~ political_stability + democracy + 
    scs + post2022, data = df)

Residuals:
    Min      1Q  Median      3Q     Max 
-42.770 -10.692   3.055  12.505  35.446 

Coefficients:
                    Estimate Std. Error t value Pr(>|t|)    
(Intercept)          34.8375    25.0269   1.392 0.172705    
political_stability   0.0058     0.3129   0.019 0.985314    
democracy             7.0189     1.7532   4.004 0.000309 ***
scs1                -14.8289     7.3357  -2.021 0.050929 .  
post20221            10.4480     6.5032   1.607 0.117129    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 20.1 on 35 degrees of freedom
Multiple R-squared:  0.3594,    Adjusted R-squared:  0.2862 
F-statistic: 4.909 on 4 and 35 DF,  p-value: 0.003005

# 7. 建立 panel data
p_df <- pdata.frame(df, index = c("country", "year"))

# 8. Fixed Effects 模型
fe_model <- plm(
  strategic_ambiguity ~ political_stability + democracy + post2022,
  data = p_df,
  effect = "individual",
  model = "within"
)

summary(fe_model)

Oneway (individual) effect Within Model

Call:
plm(formula = strategic_ambiguity ~ political_stability + democracy + 
    post2022, data = p_df, effect = "individual", model = "within")

Balanced Panel: n = 8, T = 5, N = 40

Residuals:
    Min.  1st Qu.   Median  3rd Qu.     Max. 
-23.5751  -9.4914  -1.0830   6.3004  36.1416 

Coefficients:
                    Estimate Std. Error t-value Pr(>|t|)  
political_stability -0.21692    1.68704 -0.1286  0.89858  
democracy            1.35218   12.55351  0.1077  0.91496  
post20221           10.64700    5.59222  1.9039  0.06689 .
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Total Sum of Squares:    7958.8
Residual Sum of Squares: 6935
R-Squared:      0.12863
Adj. R-Squared: -0.17184
F-statistic: 1.42698 on 3 and 29 DF, p-value: 0.25505

# 9. Random Effects 模型
re_model <- plm(
  strategic_ambiguity ~ political_stability + democracy + post2022,
  data = p_df,
  model = "random"
)

summary(re_model)

Oneway (individual) effect Random Effect Model 
   (Swamy-Arora's transformation)

Call:
plm(formula = strategic_ambiguity ~ political_stability + democracy + 
    post2022, data = p_df, model = "random")

Balanced Panel: n = 8, T = 5, N = 40

Effects:
                 var std.dev share
idiosyncratic 239.14   15.46  0.44
individual    304.36   17.45  0.56
theta: 0.6315

Residuals:
    Min.  1st Qu.   Median  3rd Qu.     Max. 
-26.2114 -11.3538  -0.6278   8.7296  35.5464 

Coefficients:
                    Estimate Std. Error z-value Pr(>|z|)  
(Intercept)         22.40334   44.87890  0.4992  0.61764  
political_stability  0.18419    0.56014  0.3288  0.74228  
democracy            5.72589    3.31872  1.7253  0.08447 .
post20221           10.15012    4.93221  2.0579  0.03960 *
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Total Sum of Squares:    9875.1
Residual Sum of Squares: 8175.2
R-Squared:      0.17214
Adj. R-Squared: 0.10315
Chisq: 7.48575 on 3 DF, p-value: 0.057926

# 10. Hausman test：比較 FE / RE
phtest(fe_model, re_model)

    Hausman Test

data:  strategic_ambiguity ~ political_stability + democracy + post2022
chisq = 0.18499, df = 3, p-value = 0.98
alternative hypothesis: one model is inconsistent

# 11. OLS + FE + RE 結果整理成表格
stargazer(
  ols_model, fe_model, re_model,
  type = "text",
  title = "Regression Results: Strategic Ambiguity",
  dep.var.labels = "Strategic Ambiguity",
  covariate.labels = c(
    "Political Stability",
    "Democracy",
    "SCS Claimant",
    "Post-2022"
  ),
  omit.stat = c("ser"),
  digits = 3
)

Regression Results: Strategic Ambiguity
=====================================================================
                                   Dependent variable:               
                    -------------------------------------------------
                                   Strategic Ambiguity               
                             OLS                     panel           
                                                    linear           
                             (1)                 (2)           (3)   
---------------------------------------------------------------------
Political Stability         0.006               -0.217        0.184  
                           (0.313)             (1.687)       (0.560) 
                                                                     
Democracy                 7.019***              1.352         5.726* 
                           (1.753)             (12.554)      (3.319) 
                                                                     
SCS Claimant              -14.829*                                   
                           (7.336)                                   
                                                                     
Post-2022                  10.448              10.647*       10.150**
                           (6.503)             (5.592)       (4.932) 
                                                                     
Constant                   34.837                             22.403 
                          (25.027)                           (44.879)
                                                                     
---------------------------------------------------------------------
Observations                 40                   40            40   
R2                          0.359               0.129         0.172  
Adjusted R2                 0.286               -0.172        0.103  
F Statistic         4.909*** (df = 4; 35) 1.427 (df = 3; 29)  7.486* 
=====================================================================
Note:                                     *p<0.1; **p<0.05; ***p<0.01

ggplot(df, aes(x = democracy, y = strategic_ambiguity, label = country)) +
  geom_point(size = 3) +
  geom_smooth(method = "lm", se = TRUE) +
  ggrepel::geom_text_repel(size = 3) +
  labs(
    title = "Democracy and Strategic Ambiguity",
    x = "Democracy",
    y = "Strategic Ambiguity"
  ) +
  theme_minimal()

`geom_smooth()` using formula = 'y ~ x'

Warning: The following aesthetics were dropped during statistical transformation: label.
ℹ This can happen when ggplot fails to infer the correct grouping structure in
  the data.
ℹ Did you forget to specify a `group` aesthetic or to convert a numerical
  variable into a factor?

df %>%
  mutate(period = ifelse(year < 2022, "Before 2022", "Post-2022")) %>%
  group_by(country, period) %>%
  summarise(mean_ambiguity = mean(strategic_ambiguity, na.rm = TRUE), .groups = "drop") %>%
  ggplot(aes(x = period, y = mean_ambiguity, group = country)) +
  geom_line() +
  geom_point(size = 3) +
  facet_wrap(~ country) +
  labs(
    title = "Changes in Strategic Ambiguity Before and After 2022",
    x = "",
    y = "Average Strategic Ambiguity"
  ) +
  theme_minimal()

library(tidyverse)

df %>%
  group_by(country) %>%
  summarise(mean_ambiguity = mean(strategic_ambiguity, na.rm = TRUE)) %>%
  arrange(mean_ambiguity) %>%
  ggplot(aes(x = reorder(country, mean_ambiguity), y = mean_ambiguity)) +
  geom_col() +
  coord_flip() +
  labs(
    title = "Average Strategic Ambiguity by Country, 2019–2023",
    x = "country",
    y = "Average Strategic Ambiguity"
  ) +
  theme_minimal()

library(tidyverse)
library(marginaleffects)

Warning: package 'marginaleffects' was built under R version 4.5.2

m_dem <- lm(
  strategic_ambiguity ~ political_stability * democracy +
    scs + post2022 + trade_china + sec_us,
  data = df
)

summary(m_dem)

Call:
lm(formula = strategic_ambiguity ~ political_stability * democracy + 
    scs + post2022 + trade_china + sec_us, data = df)

Residuals:
    Min      1Q  Median      3Q     Max 
-40.134 -11.197   4.241  10.049  31.564 

Coefficients:
                              Estimate Std. Error t value Pr(>|t|)  
(Intercept)                   104.7533   183.0311   0.572   0.5711  
political_stability            -1.1327     2.3752  -0.477   0.6367  
democracy                      -8.7921    30.6290  -0.287   0.7759  
scs1                          -17.9451     8.0802  -2.221   0.0336 *
post20221                       9.9828     6.9495   1.436   0.1606  
trade_china                    -5.9151    56.5985  -0.105   0.9174  
sec_us                         27.5880    25.4438   1.084   0.2863  
political_stability:democracy   0.2174     0.4165   0.522   0.6053  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 20.58 on 32 degrees of freedom
Multiple R-squared:  0.386, Adjusted R-squared:  0.2517 
F-statistic: 2.874 on 7 and 32 DF,  p-value: 0.01907

plot_slopes(
  m_dem,
  variables = "political_stability",
  condition = "democracy"
) +
  labs(
    title = "Marginal Effect of Political Stability across Democracy",
    x = "Democracy",
    y = "Marginal Effect of Political Stability on Strategic Ambiguity"
  ) +
  theme_minimal()