如何实现分组累计求和

boydreaming 发表于 2015-10-20 08:16
请问假如结果只是要
name fenshu
b       6

同问，请问您找到方法了吗？

1. require(dplyr)
   
2. t=data.frame(name=c("b","a","b","a","a"),fenshu=1:5)
   
3. t %>% group_by(name) %>% mutate(cum_fenshu = cumsum(fenshu))

复制代码

boydreaming 发表于 2015-10-20 08:16
请问假如结果只是要
name fenshu
b       6

1. t=data.frame(name=c("b","a","b","a","a"),fenshu=1:5)
   
2. t %>% group_by(name) %>% summarise(sum_fenshu = sum(fenshu))

复制代码

1. library(dplyr)
   
2. name=c("b","a","b","a","a")
   
3. fenshu=c(1,2,3,4,5)
   
4. df=data.frame(name,fenshu)
   
5. df %>% arrange(name) %>%
   
6.   filter(name=="a") %>%
   
7.   mutate(cum_fenshu = cumsum(fenshu))->df1
   
8. df %>% arrange(name) %>%
   
9.   filter(name=="b") %>%
   
10.   mutate(cum_fenshu = cumsum(fenshu))->df2
    
11. rbind(df1,df2)

复制代码

1. library(dplyr)
   
2. df=data.frame(name=c("b","a","b","a","a"),fenshu=1:5)
   
3. df %>% arrange(name) %>%
   
4.   group_by(name) %>%
   
5.   mutate(cum_fenshu = cumsum(fenshu))

复制代码

楼上写的真麻烦，不是分组累计求和，就一个函数的事么