R dataframe - if lag value is NA then take previous

Question

my df looks like this

df <- read.table(text="
  id           date           
  2              NA
  2      2018-11-01
  2              NA     
  2              NA 
  2      2019-11-01 
  2              NA  
  3      2018-01-01
  3              NA      
  3              NA    
  3              NA        
", header=TRUE)

My goal is to fill NAs by value above. But once is value not NA, then I'd like to fill next NAs with that. In other words, result should look like this:

result <- read.table(text="
  id           date           
  2              NA
  2      2018-11-01
  2      2018-11-01     
  2      2018-11-01
  2      2019-11-01 
  2      2019-11-01  
  3      2018-01-01
  3      2018-01-01      
  3      2018-01-01    
  3      2018-01-01        
", header=TRUE)

Note that first row for id 2 is NA and since there is no row above it, value is still NA,

score 1 · Answer 1 · answered May 24 '19 at 14:59

1

An option would be fill

library(dplyr)
df %>% 
    group_by(id) %>% 
    fill(date)

answered May 24 '19 at 14:59

akrun

874,273
37
540
662

R dataframe - if lag value is NA then take previous

1 Answers1