Creating a new date variable based on other date variables

[jdd]

Hi- I am trying to create a date variable from two other date variables. Basically, I merged two separate datasets that both had a different named date variable, but I want to merge them into one variable that has either the non-missing date or the earlier date.

Here is an the minimally reproducible example:

MRE_date1 <- as.Date(c("2001-01-01", NA, "2002-01-01", "2003-01-01"))
MRE_date2 <- as.Date(c(NA, "2003-01-01", NA, "2001-01-01"))

MRE <- data.frame(MRE_date1, MRE_date2)
MRE

So, I want to create a new variable (MRE_finaldate) that is either the non-missing date or the earlier date. In this case MRE_finaldate would be: 2001-01-01, 2003-01-01, 2002-01-01, 2001-01-01. However, my working dataset is too large to do this manually. Does anyone know how I can produce a code that does this?

Thank you!!!

[FJCC]

Here is one method. It takes advantage of the fact that the min() function will return NA if one of the values is NA.

library(tidyverse)
#> Warning: package 'tibble' was built under R version 4.1.2

MRE_date1 <- as.Date(c("2001-01-01", NA, "2002-01-01", "2003-01-01"))
MRE_date2 <- as.Date(c(NA, "2003-01-01", NA, "2001-01-01"))

MRE <- data.frame(MRE_date1, MRE_date2)
MRE
#>    MRE_date1  MRE_date2
#> 1 2001-01-01       <NA>
#> 2       <NA> 2003-01-01
#> 3 2002-01-01       <NA>
#> 4 2003-01-01 2001-01-01
MRE <- MRE |> rowwise() |> mutate(FinalDate = coalesce(min(MRE_date1, MRE_date2),
                                          MRE_date1, MRE_date2))
MRE                     
#> # A tibble: 4 x 3
#> # Rowwise: 
#>   MRE_date1  MRE_date2  FinalDate 
#>   <date>     <date>     <date>    
#> 1 2001-01-01 NA         2001-01-01
#> 2 NA         2003-01-01 2003-01-01
#> 3 2002-01-01 NA         2002-01-01
#> 4 2003-01-01 2001-01-01 2001-01-01

^{Created on 2022-03-09 by the reprex package (v2.0.1)}

[jdd]

Thank you so much! This worked!!!