Completed exercise 2.

Removed `commented_code_linter`, as it's very noisy during development.

code/2.R

@@ -0,0 +1,34 @@
+data <- read.table("data/2dtmc_data.csv", sep = "\t")
+data <- data$x
+
+occur_matrix <- matrix(
+  c(c(0, 0), c(0, 0)),
+  nrow = 2,
+  ncol = 2,
+  byrow = TRUE
+)
+
+for (data_idx in 2:length(data)) {
+  prev <- data[data_idx - 1]
+  cur <- data[data_idx]
+
+  occur_matrix[1 + cur, 1 + prev] <- occur_matrix[1 + cur, 1 + prev] + 1
+}
+
+cat("Occurrences matrix:\n")
+print(occur_matrix)
+write.table(occur_matrix, "tables/2-occur.csv", sep = "\t", col.names = FALSE, row.names = FALSE)
+
+prob_matrix <- apply(
+  occur_matrix,
+  1, function(row) row / sum(row)
+)
+prob_matrix <- apply(prob_matrix, 2, function(cell) round(cell, 2))
+
+cat("Probability matrix:\n")
+print(prob_matrix)
+write.table(prob_matrix, "tables/2-prob.csv", sep = "\t", col.names = FALSE, row.names = FALSE)
+
+alpha <- prob_matrix[1, 1]
+beta <- prob_matrix[2, 1]
+cat(sprintf("α = %f, β = %f\n", alpha, beta))