Create a new variable with value from another variable if condition is met-CodePudding

The data that I have is as:

dt_1        cond1
2009-05-04  1   
2009-07-04  0

I want to add a variable that has date from a existing variable, if cond1 is 1 else I want set it to missing.

dt_1        cond1   cond1_dt
2009-05-04  1       2009-05-04
2009-07-04  0       NA

if the new variable could retain the format that would be the best.

CodePudding user response：

Supose dat is your data frame, which has dt_1 and cond1 columns, then you can create the dt_1 column that satisfies your requirement as follows:

dat <- structure(list(dt_1 = structure(c(14368, 14429), class = "Date"), 
    cond1 = 1:0), row.names = c(NA, -2L), class = "data.frame")

In base R:

# Assign the value of cond1 column to cond1_dt as a `character date`

dat$cond1_dt <- ifelse(dat$cond1 == 1, as.character.Date(dat$dt_1), NA)

dat
        dt_1 cond1   cond1_dt
1 2009-05-04     1 2009-05-04
2 2009-07-04     0       <NA>

# Change the format from a `character date` to a `Date`
dat$cond1_dt <- as.Date(dat$cond1_dt)


dat
        dt_1 cond1   cond1_dt
1 2009-05-04     1 2009-05-04
2 2009-07-04     0       <NA>

str(dat)
'data.frame':   2 obs. of  3 variables:
 $ dt_1    : Date, format: "2009-05-04" ...
 $ cond1   : int  1 0
 $ cond1_dt: Date, format: "2009-05-04" ...

In dplyr :

dat %>% as_tibble() %>% 
        mutate(cond1_dt = if_else(cond1 == 1, dt_1, as.Date(as.character(NA)))) 
# A tibble: 2 × 3
  dt_1       cond1 cond1_dt  
  <date>     <int> <date>    
1 2009-05-04     1 2009-05-04
2 2009-07-04     0 NA