Korrelation Matrix: Cross-Border M&A Analys

CBA-transaktioner inom TMT-sektorn 1997-2023

Den här sidan presenterar en korrelationsmatris som analyserar sambanden mellan variabler i CBA-transaktioner (M&A). Plottet nedan visualiserar de parvisa korrelationerna mellan numeriska variabler. Positiva korrelationer visas i blått, negativa korrelationer i rött och neutrala värden i vitt. Cirklarnas storlek representerar styrkan av korrelationen.
Correlation Matrix Plot

R Script

Nedan är R-koden som används för att skapa korrelationsmatrisen och generera visualiseringen. Koden använder paketet `ggplot2` för att plotta och `dplyr` för datamanipulation. 
    
library(readr)
library(dplyr)
library(ggplot2)
library(reshape2)

# Load the cleaned dataset
file_path <- "Book1_cleaned.csv"
data <- read_csv(file_path)

# Convert non-numeric columns to factors (if needed)
data <- data %>% mutate(across(where(is.character), as.factor))

# Select only numeric columns
data_numeric <- data %>% select(where(is.numeric))

# Compute correlation matrix
cor_matrix <- cor(data_numeric, use = "pairwise.complete.obs")

# Convert correlation matrix to long format for ggplot
cor_melt <- melt(cor_matrix)

# Clean variable names (replace underscores with spaces)
cor_melt$Var1 <- gsub("_", " ", cor_melt$Var1)
cor_melt$Var2 <- gsub("_", " ", cor_melt$Var2)

# Create a lower triangular mask to match the reference image
cor_melt <- cor_melt[as.numeric(as.factor(cor_melt$Var1)) >= as.numeric(as.factor(cor_melt$Var2)), ]

# Plot correlation matrix with circles (no border and mirrored layout)
heatmap_plot <- ggplot(cor_melt, aes(Var2, Var1, fill = value, size = abs(value))) +
  geom_point(shape = 21, stroke = 0) +
  scale_fill_gradient2(low = "red", mid = "white", high = "blue", midpoint = 0, limits = c(-1, 1)) +
  scale_size(range = c(2, 10)) +
  labs(
    title = "Correlation Matrix",
    subtitle = "Visualization of pairwise correlations between numeric variables",
    x = "",
    y = ""
  ) +
  theme_minimal(base_size = 14) +
  theme(
    axis.text.x = element_text(angle = 45, hjust = 1, color = "black", size = 12),
    axis.text.y = element_text(color = "black", size = 12),
    axis.title.x = element_blank(),
    axis.title.y = element_blank(),
    panel.grid = element_blank(),
    plot.title = element_text(hjust = 0.5, face = "bold", size = 18),
    plot.subtitle = element_text(hjust = 0.5, size = 14, color = "gray40"),
    legend.position = "right",
    legend.text = element_text(size = 12),
    legend.title = element_blank()
  ) +
  guides(fill = guide_colorbar(title = "Correlation", barwidth = 1, barheight = 15), size = FALSE)

# Save the plot as an image
ggsave("correlation_plot.png", heatmap_plot, width = 10, height = 8, dpi = 300)

# Display plot
print(heatmap_plot)

# Optional: Save correlation matrix to a CSV file
write.csv(cor_matrix, "correlation_matrix.csv")