取R中多次分组后的百分比(dplyr)

Question

我在 R 中有以下 table：

id	var1	var2	value
ID1	A	X	1
ID2	B	X	2
ID3	C	X	3
ID4	D	X	4
ID5	A	Y	2
ID6	C	Y	5
ID7	B	Y	3

然后在 dplyr 中 group_by var1 和 var2 并取每个分组的比例结果为：

id	var1	var2	value
ID1	A	X	1/3
ID2	A	Y	2/3
ID3	C	X	3/8
ID4	C	Y	5/8
ID5	B	X	2/5
ID6	B	Y	3/5
ID7	D	X	1

我试过了：

id = c("ID1","ID2","ID3","ID4","ID5","ID6","ID7")
var1 = c("A","B","C","D","A","C","B")
var2 = c(rep("X",4),rep("Y",3))
value = c(1,2,3,4,2,5,3)
data = data.frame(id,var1,var2,value);data
library(dplyr)
data%>%
  group_by(var1,var2)%>%
  summarise(prop = sum(value))

但它只对 var1 和 var2 进行分组。有帮助吗？

Answer 1

这可能有效

library(dplyr)

data %>%
  group_by(var1)%>%
  mutate(value = value/sum(value)) %>%
  arrange(var1, var2)

  id    var1  var2  value
  <chr> <chr> <chr> <dbl>
1 ID1   A     X     0.333
2 ID5   A     Y     0.667
3 ID2   B     X     0.4  
4 ID7   B     Y     0.6  
5 ID3   C     X     0.375
6 ID6   C     Y     0.625
7 ID4   D     X     1

Answer 2

data.table

library(data.table)
setDT(df)[, res := proportions(value), by = var1][order(var1)]

基础

df$res <- ave(df$value, list(df$var1), FUN = proportions)

取R中多次分组后的百分比(dplyr)

Take the percentage after multiple grouping in R (dplyr)

grouping

r

percentage

dplyr