Correction of the inner join for players playing for 2 clubs in one season

PreparingData.R

@@ -63,11 +63,17 @@ rowsPIn2CGlobal <- c()
 for (i in 1:length(allLaLigaPlayer)) {
   repeatedPlayers <- which(duplicated(allLaLigaPlayer[[i]]$player) == TRUE)
   playersIn2C <- allLaLigaPlayer[[i]]$player[repeatedPlayers]
+  nacionalitiesIn2C <- allLaLigaPlayer[[i]]$nationality[repeatedPlayers]
+  birthyearssIn2C <- allLaLigaPlayer[[i]]$birth_year[repeatedPlayers]
   season <- switch (i, "17-18", "18-19", "19-20")
-  for (name in playersIn2C) {
-    rows <- which(allLaLigaPlayer[[i]]$player == name)
+  for (j in 1:length(playersIn2C)) {
+    rows <- which(allLaLigaPlayer[[i]]$player == playersIn2C[j] & 
+                    allLaLigaPlayer[[i]]$nationality == nacionalitiesIn2C[j] & 
+                    allLaLigaPlayer[[i]]$birth_year == birthyearssIn2C[j])
+    if(length(rows) > 1) {
       teams <- c(allLaLigaPlayer[[i]]$squad[rows[1]], allLaLigaPlayer[[i]]$squad[rows[2]])
-    rowsPIn2CGlobal <- rbind(rowsPIn2CGlobal, c(name, rows, teams, season))
+      rowsPIn2CGlobal <- rbind(rowsPIn2CGlobal, c(playersIn2C[j], rows, teams, season))
+    }
   }
 }
 colnames(rowsPIn2CGlobal) <- c("Jugador", "Fila 1", "Fila 2", "Equipo 1º", "Equipo 2º", "Temporada")
@@ -148,9 +154,13 @@ laLigaPlayers.1920 <- rbind(laLigaPlayers.1920, players)
 ##############
 # INNER JOIN #
 ##############
-laLigaPlayers.1719 <- inner_join(laLigaPlayers.1718, laLigaPlayers.1819, by = c("player" = "player"), suffix = c(".1718", ".1819"))
-auxLaLigaPlayers.1920 <- inner_join(laLigaPlayers.1920, laLigaPlayers.1920, by = c("player" = "player"), suffix = c(".1920", ".1920"))
-laLigaPlayers <- inner_join(laLigaPlayers.1719, auxLaLigaPlayers.1920, by = c("player" = "player"))
+laLigaPlayers <- inner_join(laLigaPlayers.1718, laLigaPlayers.1819, 
+                                 by = c("player", "nationality", "birth_year"), 
+                                 suffix = c("", ".1819")) %>%
+                 inner_join(laLigaPlayers.1920, 
+                                 by = c("player", "nationality", "birth_year"), 
+                                 suffix = c(".1718", ".1920"))
+
 
 # Veamos que jugadores jugaron en 2 CLUBES en algunas de las 3 temporadas contempladas
 repeatedPlayers <- which(duplicated(laLigaPlayers$player) == TRUE)
@@ -167,6 +177,6 @@ rowsPIn2C
 # CONJUNTO de ENTRENAMIENTO # # CONJUNTO de TEST #
 ############################# ####################
 set.seed(5682)
-ind <- sample(2, nrow(laLigaPlayers), replace = TRUE, prob = c(0.8, 0.2))
+ind <- sample(2, nrow(laLigaPlayers), replace = FALSE, prob = c(0.8, 0.2))
 train <- laLigaPlayers[ind == 1, ]
 test <- laLigaPlayers[ind == 2, ]