R 中多個因素的匯總統計？

Question

對於每款游戲，我都有如下所示的游戲數據：

ID	位置	團隊	歐普	分數
0	一個	鴨子	青蛙	2
1	乙	鴨子	青蛙	15
2	乙	鴨子	青蛙	20
3	C	鴨子	青蛙	7
4	C	鴨子	青蛙	9.5
5	C	鴨子	青蛙	10
6	一個	青蛙	鴨子	3
7	一個	青蛙	鴨子	0.5
8	乙	青蛙	鴨子	17
9	乙	青蛙	鴨子	13
10	乙	青蛙	鴨子	21
11	C	青蛙	鴨子	8.5

我想獲得每個 position 和團隊（以及他們的對手）的平均分數，所以結果看起來像這樣。

ID	位置	團隊	歐普	分數	Team_A_Avg	Opp_A_Avg	Team_B_Avg	Opp_B_Avg	Team_C_Avg	Opp_C_Avg
0	一個	鴨子	青蛙	2	2	1.75	17.5	17	8.8333	8.5
1	乙	鴨子	青蛙	15	2	1.75	17.5	17	8.8333	8.5
2	乙	鴨子	青蛙	20	2	1.75	17.5	17	8.8333	8.5
3	C	鴨子	青蛙	7	2	1.75	17.5	17	8.8333	8.5
4	C	鴨子	青蛙	9.5	2	1.75	17.5	17	8.8333	8.5
5	C	鴨子	青蛙	10	2	1.75	17.5	17	8.8333	8.5
6	一個	青蛙	鴨子	3	1.75	2	17	17.5	8.5	8.8333
7	一個	青蛙	鴨子	0.5	1.75	2	17	17.5	8.5	8.8333
8	乙	青蛙	鴨子	17	1.75	2	17	17.5	8.5	8.8333
9	乙	青蛙	鴨子	13	1.75	2	17	17.5	8.5	8.8333
10	乙	青蛙	鴨子	21	1.75	2	17	17.5	8.5	8.8333
11	C	青蛙	鴨子	8.5	1.75	2	17	17.5	8.5	8.8333

go 關於這個問題的最佳方法是什么？

Answer 1

正如@Maurits Evers 所評論的那樣，您展示 output 的方式並沒有真正的意義。 似乎您想要一個單獨的 output，每個團隊的平均得分和 position。 另外，你只給了我們每一行的分數，我認為這是Team的分數，所以我們沒有對手的分數來計算平均值。 我會使用dplyr summarise function。

這是您的數據：

game = data.frame(id = c(0:11),
                  Pos = c("A", "B", "B", "C", "C", "C","A","A", "B", "B", "B","C"),
                  Team = c("Duck","Duck","Duck","Duck","Duck","Duck","Frog","Frog","Frog","Frog","Frog","Frog"),
                  Opp = c("Frog","Frog","Frog","Frog","Frog","Frog","Duck","Duck","Duck","Duck","Duck","Duck"),
                  Score = c(2, 15, 20, 7, 9.5, 10, 3, 0.5, 17, 13, 21, 8.5))

首先是 position 的平均值：

library(dplyr)
Pos_av = game%>%    #creat a new data.frame called "Pos_av" which is taking data from "game" and piping it (%>%) into different functions 
  group_by(Pos)%>%    #first into a grouping function so we chose the variable we want to find the average for
  summarise(Pos_Mean = mean(Score))    # the we pipe into summarise function where we name our new variable (Pos_Mean) and then define the function we want to use to summarise it (in this case the mean)

這給了我們：

然后對於團隊來說也是一樣的意思：

Team_av = game%>%
  group_by(Team)%>%
  summarise(Team_Mean = mean(Score))

要獲得每個團隊和 position 的平均值，請按兩個變量分組：

Both_av = game%>%
  group_by(Team, Pos)%>%
  summarise(Mean = mean(Score))

Answer 2

您可以通過循環遍歷 dataframe 和所有條件來設置每個單元格的值，具體取決於此條件和此對手/團隊的平均值：

## The name of the variable holding the data.frame is "df"

## Expand the dataframe to contain your desired variables
for(t in c("Team","Opp")){
  for(p in c("A","B","C")){
    df[[paste(t,"_",p,"_","Avg",sep="")]]=NA
  }
}

## Loop through the data to compute the means
for(i in 1:dim(df)[1]){
  for(t in c("Team","Opp")){
    for(p in c("A","B","C")){
      ## For each case i, each Team t, and each Position p, compute the mean and store it:
      df[[paste(t,"_",p,"_","Avg",sep="")]][i] = mean(df$Score[df$Team==df[[t]][i] & df$Pos==p])
    }
  }
}

這導致 dataframe：

> df
   Id Pos Team  Opp Score Team_A_Avg Team_B_Avg Team_C_Avg Opp_A_Avg Opp_B_Avg Opp_C_Avg
1   0   A Duck Frog   2.0       2.00       17.5   8.833333      1.75      17.0  8.500000
2   1   B Duck Frog  15.0       2.00       17.5   8.833333      1.75      17.0  8.500000
3   2   B Duck Frog  20.0       2.00       17.5   8.833333      1.75      17.0  8.500000
4   3   C Duck Frog   7.0       2.00       17.5   8.833333      1.75      17.0  8.500000
5   4   C Duck Frog   9.5       2.00       17.5   8.833333      1.75      17.0  8.500000
6   5   C Duck Frog  10.0       2.00       17.5   8.833333      1.75      17.0  8.500000
7   6   A Frog Duck   3.0       1.75       17.0   8.500000      2.00      17.5  8.833333
8   7   A Frog Duck   0.5       1.75       17.0   8.500000      2.00      17.5  8.833333
9   8   B Frog Duck  17.0       1.75       17.0   8.500000      2.00      17.5  8.833333
10  9   B Frog Duck  13.0       1.75       17.0   8.500000      2.00      17.5  8.833333
11 10   B Frog Duck  21.0       1.75       17.0   8.500000      2.00      17.5  8.833333
12 11   C Frog Duck   8.5       1.75       17.0   8.500000      2.00      17.5  8.833333