Scatter corr plot with new colour argument and 1st elimination of correlated variables

BivariateAnalysis.R

@@ -23,7 +23,7 @@ puntos2d <- function(variableEnX, variableEnY, datos, temporada, color) {
 }
 
 # Scatter plot for 2 variables with their correlation and regression line
-scatterForCorr <- function(datos, variableX, variableY, season) {
+scatterForCorr <- function(datos, variableX, variableY, season, col) {
   if(missing(season)) {
     X <- paste(variableX, sep = "")
     Y <- paste(variableY, sep = "")
@@ -37,10 +37,11 @@ scatterForCorr <- function(datos, variableX, variableY, season) {
   
   sp <- ggscatter(datos, x = X, y = Y,
                   add = "reg.line",  # Add regressin line
-                  add.params = list(color = "#00688B", fill = "lightgray"), # Customize reg. line
+                  add.params = list(color = col, fill = "lightgray"), # Customize reg. line
                   conf.int = TRUE # Add confidence interval
   )
-  sp + stat_cor(method = "pearson", label.x = medianX, label.y = maxY+10)
+  sp + stat_cor(method = "pearson", label.x = medianX, label.y = maxY+1.5) + 
+    xlab(titleForYAxis(variableX)) + ylab(titleForYAxis(variableY)) 
 }
 
 #############
@@ -152,21 +153,26 @@ sp + stat_cor(method = "pearson", label.x = , label.y = 30)
 ######################################
 # SELECCION DE LAS VARIABLES FINALES #
 ######################################
+# 1era SELECCION
 # Quitamos: "passes_completed.", "passes_received.", "shots_on_target.", "dribbles_completed.",
 # "tackles_won","aerials_won.", "pens_made.", "pens_saved.", "pens_allowed.",
 variables <- c("xa.", "passes.", "passes_pct.", "passes_total_distance.", "assisted_shots.", "passes_switches.", 
                "pass_targets.", "passes_received_pct.", "passes_pressure.",
-               "xg.", "npxg.",  "shots_total.", "shots_on_target_pct.", "goals_per_shot.",  "pens_att.", "pens_made_pct.", 
+               "xg.", "npxg.",  "shots_total.", "shots_on_target_pct.", "pens_att.", "pens_made_pct.", 
                "dribbles.",  "dribbles_completed_pct.", "goals_assists_per90.",
                "sca_passes_dead.", "gca_passes_dead.", "sca_passes_live.", "gca_passes_live.", "sca_dribbles.", "gca_dribbles.", "sca_fouled.", "gca_fouled.",
                "passes_intercepted.", "ball_recoveries.", "pressure_regains.", "fouls.",  "tackles.", "tackles_pct.", "aerials_contested.", "aerials_won_pct.", 
                "goals_against_per90_gk.", "pens_played.", "pens_saved_pct.", "shots_on_target_against.", "saves.", "save_pct.")
-
-laLigaPlayersStudy <- laLigaPlayers %>% select(all_off(variables))
+globalColNames <- c()
+for (variable in variables) {
+  variable <- gsub("\\.", "", variable)
+  globalColNames <- c(globalColNames, variable)
+}
+laLigaPlayersStudy <- laLigaPlayersStudy %>% select(all_of(globalColNames))
 laLigaPlayersStudy <- data.frame(lapply(laLigaPlayersStudy, as.numeric))
-LaLigaPlayersStudyStand <- as.data.frame(scale(laLigaPlayersStudy))
-correlacion <- round(cor(LaLigaPlayersStudyStand), 1)
-p.mat <- cor_pmat(LaLigaPlayersStudyStand)
+laLigaPlayersStudyStand <- data.frame(scale(laLigaPlayersStudy))
+correlacion <- round(cor(laLigaPlayersStudyStand), 1)
+p.mat <- cor_pmat(laLigaPlayersStudyStand)
 mypalette <- brewer.pal(n = 3, name = 'BuPu')
 ggcorrplot(correlacion, type = "upper",
            title = "Matriz de correlaciones ordenada",
@@ -174,9 +180,30 @@ ggcorrplot(correlacion, type = "upper",
            pch.cex = 1.2, lab_size = 2.5, tl.cex = 9,
            colors = mypalette, ggtheme = ggplot2::theme_gray, p.mat = p.mat, insig = "blank")
 
+###########
+# NIVEL 1 #
+###########
+scatterForCorr(laLigaPlayersStudy, "passes_total_distance", "passes", col = "#9A32CD")
+scatterForCorr(laLigaPlayersStudy, "npxg", "xg", col = "#CDC673")
+scatterForCorr(laLigaPlayersStudy, "sca_passes_live", "assisted_shots", col = "#9A32CD")
+scatterForCorr(laLigaPlayersStudy, "save_pct", "goals_against_per90_gk", col = "#40E0D0")
+scatterForCorr(laLigaPlayersStudy, "saves", "shots_on_target_against", col = "#40E0D0")
+
+# 2nda SELECCION
+# Quitamos: "passes_total_distance.", "xg.", "assisted_shots.", "goals_against_per90_gk.", "shots_on_target_against.", 
+variables <- c("xa.", "passes.", "passes_pct.",  "passes_switches.", 
+               "pass_targets.", "passes_received_pct.", "passes_pressure.",
+               "npxg.",  "shots_total.", "shots_on_target_pct.", "pens_att.", "pens_made_pct.", 
+               "dribbles.",  "dribbles_completed_pct.", "goals_assists_per90.",
+               "sca_passes_dead.", "gca_passes_dead.", "sca_passes_live.", "gca_passes_live.", "sca_dribbles.", "gca_dribbles.", "sca_fouled.", "gca_fouled.",
+               "passes_intercepted.", "ball_recoveries.", "pressure_regains.", "fouls.",  "tackles.", "tackles_pct.", "aerials_contested.", "aerials_won_pct.", 
+               "pens_played.", "pens_saved_pct.", "saves.", "save_pct.")
 # Paradas contra Tiros a puerta en contra
 goalkeepers <- laLigaPlayers[which(laLigaPlayers$position.1718 == "GK"),]
 datos <- laLigaPlayers[which(laLigaPlayers$position.1718 == "GK")]
 puntos2d("saves", 
          "shots_on_target_against", 
          goalkeepers, "1718", "firebrick3")
+
+
+