如何在R中找到数据框的概率？

Question

我有一个数据框。这是其中的一小部分：

     SubjectId Gender Groups  ExtraCalories        GW
1:         1      F     G3    -1310.00000    0.000000
2:         2      M     G6     -920.79656    4.331278
3:         3      M     G2      -25.39517    4.727376
4:         4      M     G5      169.25645    3.543941
5:         5      M     G5     -340.67235    4.591774
---                                               
996:     996      F     G1     464.82543     5.933792
997:     997      M     G8    -323.65136     5.024453
998:     998      F     G3      77.92138     5.383686
999:     999      M     G9    -237.83700     5.423941
1000:   1000      F     G9    -400.44831     6.837965

我如何找到女性选择 G5 的概率。

Answer 1

这是一个基本 R 解决方案，其中 prop.table 计数为 table。

prop.table(table(df1[c('Gender', 'Groups')]))
#      Groups
#Gender  G1  G2  G3  G5  G6  G8  G9
#     F 0.1 0.0 0.2 0.0 0.0 0.0 0.1
#     M 0.0 0.1 0.0 0.2 0.1 0.1 0.1

dput 格式的数据

df1 <-
structure(list(SubjectId = c(1L, 2L, 3L, 4L, 5L, 996L, 997L, 
998L, 999L, 1000L), Gender = c("F", "M", "M", "M", "M", "F", 
"M", "F", "M", "F"), Groups = c("G3", "G6", "G2", "G5", "G5", 
"G1", "G8", "G3", "G9", "G9"), ExtraCalories = c(-1310, -920.79656, 
-25.39517, 169.25645, -340.67235, 464.82543, -323.65136, 77.92138, 
-237.837, -400.44831), GW = c(0, 4.331278, 4.727376, 3.543941, 
4.591774, 5.933792, 5.024453, 5.383686, 5.423941, 6.837965)), 
class = "data.frame", row.names = c("1:", "2:", "3:", "4:", "5:", 
"996:", "997:", "998:", "999:", "1000:"))

Answer 2

我猜你想要按频率近似概率。最后两个选项比Base R Solution P>更为常用

base R

nrow(df[df$Gender == "F" & df$Groups == "G5",])/nrow(df[df$Gender == "F",])

dplyr

library(dplyr)
df %>% filter(Gender == "F") %>%
   group_by(Groups) %>%
   summarise(n = n()) %>%
   ungroup() %>%
   mutate(p = n/n())   

数据表

library(data.table)
setDT(df)
df[Gender == "F"][,.(n = .N),by = Groups][,.(n/.N)]