在 dplyr 中按组创建具有最近日期的新变量

Question

我有一个数据框，我想用Id 为该组创建 2 个新变量。 首先我需要按Id 分组并通过createdDate 获取最近的日期然后我需要根据最近的日期再次获取Lead_DataSource__c

这是我的数据框的尾部；

tail(df)


Id                  CreatedDate Lead_DataSource__c   StageName
0011000001XW3YZAA1  2020-07-17            Walk in   Quotation
0011000001XW3Z8AAL  2020-07-17            Walk in   Quotation
0011000001XW3zHAAT  2020-07-17            Walk in    Assigned
0011000001XW3zlAAD  2020-07-17            Walk in   Quotation
0011000001XW3zvAAD  2020-07-17            Walk in Closed Lost
0011000001XW3zvAAD  2020-07-17            Website Closed Lost

这是我的代码：

df_new<-df %>% group_by(Id)%>%
 mutate(numberoflead=length(Id)) %>% #number of lead
  mutate(lastcreateddateoflead=max(CreatedDate)) %>%#last date of lead
  mutate(lasttouch =max(CreatedDate)[Lead_DataSource__c])%>% #last touch

当我运行这些代码时，我没有收到任何错误，它似乎适用于 numberofleads 和 lastcreateddateoflead，但它似乎不适用于获取 lasttouch

任何人都可以帮助我了解我在这里缺少的东西吗？

Answer 1

你的问题是你使用mutate，而你应该使用summarize。然后你需要加入原来的df才能得到lasttouch。如果您在联接中添加select，则无需重命名或选择任何内容即可获得lasttouch 列。

library(dplyr)

df %>%
  group_by(Id) %>%
  summarize(numberoflead = n(),
            lastcreateddateoflead=max(CreatedDate)) %>%
  inner_join(df %>% 
               select(Id, CreatedDate, lasttouch = Lead_DataSource__c),
             by = c("Id" = "Id", "lastcreateddateoflead" = "CreatedDate"))
            
`summarise()` ungrouping output (override with `.groups` argument)
# A tibble: 6 x 4
  Id                 numberoflead lastcreateddateoflead lasttouch
  <chr>                     <int> <date>                <chr>    
1 0011000001XW3YZAA1            1 2020-07-17            Walk in  
2 0011000001XW3Z8AAL            1 2020-07-17            Walk in  
3 0011000001XW3zHAAT            1 2020-07-17            Walk in  
4 0011000001XW3zlAAD            1 2020-07-17            Walk in  
5 0011000001XW3zvAAD            2 2020-07-17            Walk in  
6 0011000001XW3zvAAD            2 2020-07-17            Website  

如果您想保留所有行（而不仅仅是每个 Id 一个摘要），请使用您的 mutate 而不是我的 summary。

df %>%
  group_by(Id) %>%
  mutate(numberoflead = n(),
            lastcreateddateoflead=max(CreatedDate)) %>%
  inner_join(df %>% 
               select(Id, CreatedDate, lasttouch = Lead_DataSource__c),
             by = c("Id" = "Id", "lastcreateddateoflead" = "CreatedDate"))

# A tibble: 8 x 7
# Groups:   Id [5]
  Id             CreatedDate Lead_DataSource_~ StageName  numberoflead lastcreateddateofl~ lasttouch
  <chr>          <date>      <chr>             <chr>             <int> <date>              <chr>    
1 0011000001XW3~ 2020-07-17  Walk in           Quotation             1 2020-07-17          Walk in  
2 0011000001XW3~ 2020-07-17  Walk in           Quotation             1 2020-07-17          Walk in  
3 0011000001XW3~ 2020-07-17  Walk in           Assigned              1 2020-07-17          Walk in  
4 0011000001XW3~ 2020-07-17  Walk in           Quotation             1 2020-07-17          Walk in  
5 0011000001XW3~ 2020-07-17  Walk in           Closed Lo~            2 2020-07-17          Walk in  
6 0011000001XW3~ 2020-07-17  Walk in           Closed Lo~            2 2020-07-17          Website  
7 0011000001XW3~ 2020-07-17  Website           Closed Lo~            2 2020-07-17          Walk in  
8 0011000001XW3~ 2020-07-17  Website           Closed Lo~            2 2020-07-17          Website 

Answer 2

嘿，我想我明白了你想要做的事情，但它可能表述得不太恰当。我相信您想要每个Id 的最大日期，然后是每个Lead_DataSource__c 的最大日期，如果您想要这样做，可以尝试：

df_new <- df %>% group_by(Id) %>%
 mutate(numberoflead=length(Id)) %>% #number of lead
 mutate(lastcreateddateoflead=max(CreatedDate)) %>%
 group_by(Lead_DataSource__c) %>% 
 mutate(lasttouch =max(CreatedDate)) %>%
 ungroup()

让我知道这是否是你想要完成的！