r - 表示每个组内有超过 1 列的组索引

Question

我有一个变量，我想获得每个组内的均值，其中为列中的每个观察列出组，我有很多这样的列。然后我想将组均值与适当的观察相关联，这样如果我从 m obs x n 不同分组的矩阵开始，我将获得 m x n 均值矩阵。例如:

> var <- round(runif(10),digits=2)

> var
[1] 0.47 0.21 0.80 0.65 0.32 0.72 0.29 0.93 0.77 0.64
> groupings <- cbind(sample(c(1,2,3), 10, replace=TRUE),
           sample(c(1,2,3), 10, replace=TRUE),
           sample(c(1,2,3,5), 10, replace=TRUE))
> groupings
      [,1] [,2] [,3]
 [1,]    3    1    5
 [2,]    1    1    5
 [3,]    2    1    5
 [4,]    3    2    3
 [5,]    2    3    1
 [6,]    1    1    1
 [7,]    2    3    1
 [8,]    1    2    1
 [9,]    3    1    5
[10,]    1    3    2

我可以通过以下(例如)分别获得每个组内的手段

> means.1 <- sapply(split(var, groupings[,1]), function(x) mean(x))
> means.2 <- sapply(split(var, groupings[,2]), function(x) mean(x))
> means.3 <- sapply(split(var, groupings[,3]), function(x) mean(x))

> means.1
    1     2     3 
0.625 0.470 0.630 
> means.2
    1         2         3 
0.5940000 0.7900000 0.4166667 
> means.3
    1      2      3      5 
0.5650 0.6400 0.6500 0.5625 

但这些单独的调用不仅效率低下，而且仍然无法得到我想要的，如下所示

       [,1]      [,2]   [,3]
[1,]  0.630 0.5940000 0.5625
[2,]  0.625 0.5940000 0.5625
[3,]  0.470 0.5940000 0.5625
[4,]  0.630 0.7900000 0.6500
[5,]  0.470 0.4166667 0.5650
[6,]  0.625 0.5940000 0.5650
[7,]  0.470 0.4166667 0.5650
[8,]  0.625 0.7900000 0.5650
[9,]  0.630 0.5940000 0.5625
[10,] 0.625 0.4166667 0.6400

Answer 1

library(dplyr)
set.seed(1000)
var <- round(runif(10),digits=2)

groupings <- cbind(sample(c(1,2,3), 10, replace=TRUE),
                     sample(c(1,2,3), 10, replace=TRUE),
                     sample(c(1,2,3,5), 10, replace=TRUE), var)

df = data.frame(groupings)
df %>% 
  group_by(V1)%>% mutate(x1 =mean(var))%>% ungroup(V1) %>% 
  group_by(V2) %>% mutate(x2=mean(var)) %>% ungroup(V2) %>% 
  group_by(V3) %>% mutate(x3=mean(var)) %>% ungroup(V3)

#      V1    V2    V3   var        x1    x2    x3
#   <dbl> <dbl> <dbl> <dbl>     <dbl> <dbl> <dbl>
#1      2     1     3  0.33 0.4775000 0.322 0.250
#2      3     3     1  0.76 0.6566667 0.470 0.750
#3      1     1     3  0.11 0.1333333 0.322 0.250
#4      3     1     5  0.69 0.6566667 0.322 0.635
#5      3     2     3  0.52 0.6566667 0.630 0.250
#6      1     3     3  0.07 0.1333333 0.470 0.250
#7      2     2     1  0.74 0.4775000 0.630 0.750
#8      2     3     5  0.58 0.4775000 0.470 0.635
#9      1     1     3  0.22 0.1333333 0.322 0.250
#10     2     1     2  0.26 0.4775000 0.322 0.260

# you can simply subset the columns