Return 按组列中最常见的值，用该值替换该列中的空值

Question

我想用分组中最常见的值替换 df 列中的 na 值

#Ex:

df <- data.frame(Home_Abbr = c('PHI', 'PHI', 'DAL', 'PHI'),
                 Home_City = c('Philadelphia', 'Philadelphia', 'Dallas', NULL))

#Desired Result

Home_Abbr   Home_City

PHI         Philadelphia
PHI         Philadelphia
DAL         Dallas
PHI         Philadelphia

这是我到目前为止尝试过的方法：

df <- df %>%
  group_by(Home_Abbr) %>%
  mutate(Home_City = names(which.max(table(Home_City))))

但是当我运行这样做时，我得到一个“无法合并 NULL 和非 NULL 结果”的错误。

Answer 1

我们可以使用函数

 Mode <- function(x) {
  ux <- unique(x)
  ux[which.max(tabulate(match(x, ux)))]
}

然后是replace

library(dplyr)
df %>% 
  group_by(Home_Abbr) %>%
  mutate(Home_City = replace(Home_City, is.na(Home_City), 
      Mode(Home_City))) %>%
  ungroup

-输出

# A tibble: 4 × 2
  Home_Abbr Home_City   
  <chr>     <chr>       
1 PHI       Philadelphia
2 PHI       Philadelphia
3 DAL       Dallas      
4 PHI       Philadelphia

数据

df <- structure(list(Home_Abbr = c("PHI", "PHI", "DAL", "PHI"), Home_City = c("Philadelphia", 
"Philadelphia", "Dallas", NA)), class = "data.frame", row.names = c(NA, 
-4L))

Return 按组列中最常见的值，用该值替换该列中的空值

Return most common value in column by group, replace null in that column with that value

r

dplyr

数据