将摘要“n”从一个数据帧添加到另一个数据帧（tidyverse）

Question

我想知道是否有办法将DATA2 中的fpc 列替换为从DATA1 获得的相应fpc？

library(tidyverse)

dat <- read.csv('https://raw.githubusercontent.com/rnorouzian/d/master/su.csv')

## 10000 rows ################
 DATA1 <- dat %>%
    group_by(across(all_of(c("gender", "pre")))) %>%
    summarise(n = n(), .groups = 'drop') %>%
    mutate(fpc = n/sum(n)) %>% 
    right_join(dat)


dat2 <- read.csv('https://raw.githubusercontent.com/rnorouzian/d/master/out.csv')
  
## 200 rows #################
DATA2 <- dat2 %>%
  group_by(across(all_of(c("gender", "pre")))) %>%
  summarise(n = n(), .groups = 'drop') %>%
  mutate(fpc = n/sum(n)) %>% 
  right_join(dat2)

Answer 1

您可以加入数据框并使用coalesce 从DATA2 中选择fpc。

library(dplyr)

result <- DATA2 %>%
           left_join(DATA1 %>% distinct(gender, pre, fpc), 
                     by = c('gender', 'pre')) %>%
           mutate(fpc = coalesce(fpc.y, fpc.x)) %>%
           select(names(DATA2))

nrow(result)
#[1] 200

Answer 2

在data.table这样做会更有效

library(data.table)
setDT(DATA2)[as.data.table(unique(DATA1[c('gender', 'pre', 'fpc')])),
            fpc := i.fpc, on = .(gender, pre)]