Replacing NA values of a column by the value of the top row of the same column of a dataframe

Question

I have the data frame below, and I want to create a script to replace the NA by the value of the top row of the same column.

DADOS <- data.frame(
  a = c(1, 2, NA, 3, 4), 
  b = c(5, 6, 7, NA, NA)
)

Below the result as you would like:

DADOSRESULTADO <- data.frame(
  a = c(1, 2, 2, 3, 4), 
  b = c(5, 6, 7, 7, 7)
)

Answer 1

The function na.locf package zoo does exactly what is requested:

library(zoo)
#> 
#> Attaching package: 'zoo'
#> The following objects are masked from 'package:base':
#> 
#>     as.Date, as.Date.numeric

na.locf(DADOS)
#>   a b
#> 1 1 5
#> 2 2 6
#> 3 2 7
#> 4 3 7
#> 5 4 7

^{Created on 2021-06-02 by the reprex package (v2.0.0)}

Answer 2

An alternative would be to use Fill package tidyr:

library(tidyr)

DADOS <- data.frame(
  a = c(1, 2, NA, 3, 4), 
  b = c(5, 6, 7, NA, NA)
)

DADOS %>% fill(a,b)

Exit:

  a b
1 1 5
2 2 6
3 2 7
4 3 7
5 4 7

Answer 3

One more option, the function data.table::nafill. The variation setnafill can be used to modify by reference:

library(data.table)

setnafill(DADOS, "locf")

DADOS
#>   a b
#> 1 1 5
#> 2 2 6
#> 3 2 7
#> 4 3 7
#> 5 4 7

You can also use "const" to fill in with a fixed value and "nocb" to use the following value. The option cols allows defining columns by name or number (default is to apply to all).