Group_by 多列并汇总唯一列

Question

我在下面有一个数据集

family	type	inc	name
AA	success	30000	Bill
AA	ERROR	15000	Bess
CC	Pending	22000	Art
CC	Pending	18000	Amy
AA	Serve not respnding d	25000	Paul
ZZ	Success	50000	Pat
ZZ	Processing	50000	Pat

我想按多列分组

下面是我的代码

df<-df1%>%
group_by(Family , type)%>%
  summarise(Transaction_count = n(), Face_value = sum(Inc))%>%
  mutate(Pct = Transaction_count/sum(Transaction_count))

我想要的是任何有相同观察的家庭，它应该只选择一个

如下图所示。

谢谢

Answer 1

您可以使用duplicated 将重复值替换为空白值。

library(dplyr)

df %>%
  group_by(family , type)%>%
  summarise(Transaction_count = n(), Face_value = sum(inc))%>%
  mutate(Pct = Transaction_count/sum(Transaction_count), 
         family = replace(family, duplicated(family), '')) %>%
  ungroup

#   family type                     Transaction_count Face_value   Pct
#  <chr>  <chr>                                <int>      <int> <dbl>
#1 "AA"   ERROR                                    1      15000 0.333
#2 ""     Serve not respnding    d                 1      25000 0.333
#3 ""     success                                  1      30000 0.333
#4 "CC"   Pending                                  2      40000 1    
#5 "ZZ"   Processing                               1      50000 0.5  
#6 ""     Success                                  1      50000 0.5  

如果您需要用于显示目的的数据，您可以查看 formattable、kable 等软件包。

数据

如果您在reproducible format 中提供数据会更容易提供帮助

df <- structure(list(family = c("AA", "AA", "CC", "CC", "AA", "ZZ", 
"ZZ"), type = c("success", "ERROR", "Pending", "Pending", "Serve not respnding    d", 
"Success", "Processing"), inc = c(30000L, 15000L, 22000L, 18000L, 
25000L, 50000L, 50000L), name = c("Bill", "Bess", "Art", "Amy", 
"Paul", "Pat", "Pat")), row.names = c(NA, -7L), class = "data.frame")