Calculate date of birth from age and event date

Question

I have two columns, one with age e.g. (34) and another column with date of the event e.g. (2019-04-26:01:20:51). I would like to create a new column that returns the date of birth based on the above two columns). Many thanks in advance for the help.

Hello Angela, welcome to SO. Could you provide a sample data for us to work with. You can get this using `dput()`. Also what have you tried to solve this problem? — snair.stack, Mar 21 '19 at 11:55
You want package `lubridaet`, function `years`: try `DateEvent - lubridate::years(34)`. — Rui Barradas, Mar 21 '19 at 15:56

snair.stack · Accepted Answer · 2019-03-21T15:20:47.890

Since there is no sample data available, I created a sample data frame using the data provided. Code snippet is given below. You don't need to use any external package for this. Instead as.POSIXlt should be enough.

df <- data.frame(event = c("2019-04-26 01:20:51"), age = c(34)) 
df$event <- as.POSIXlt(x = df$event, format = "%Y-%m-%d %H:%M:%S") # define format here    
#df$approx_DOB <- (df$event$year+1900)-df$age # 1900 to get actual year
df$approx_DOB <- NA
df$approx_DOB <- df$event
df$approx_DOB$year <- (df$event$year) - df$age # no 1900, since editing the age directly
df$YearOfBirth <- NA
df$YearOfBirth <- (df$event$year+1900) - df$age # Gives year alone

Ouput:

    > df
                event age          approx_DOB YearOfBirth
1 2019-04-26 01:20:51  34 1985-04-26 01:20:51        1985

Bonus: You can further access the elements of a POSIXlt object using $ and specifying the type required (eg: year, mon, mday etc). Then accordingly can format the approx_DOB column. Check this answer for more info.

yes, this is what I thought as well - you can calculate the approx DOB rather than exact one — GaB, Mar 21 '19 at 12:21

score 0 · Answer 2 · answered Mar 21 '19 at 12:57

Here it is another example, with tidyverse and lubridate. I believe it is a better solution since I will calculate the date of birth taking the year only and with big data sets you the calculations are quicker. The active full date won't give you the exact date of birth and probably some huge issues will come out. Thus, here it is my solution:

library(tidyverse)
library(lubridate)

df <- tibble::tibble(event = c("2018-04-26 02:30:10"), age = c(34))

df_separate <- df %>%
  dplyr::mutate(year = as.numeric(lubridate::year(event)), 
                DOB_Y_approximated = year - age)

And you get the exact year of birth, which I assume it is a better output.

  # A tibble: 1 x 4
  event                 age  year DOB_Y_approximated
  <chr>               <dbl> <dbl>              <dbl>
1 2018-04-26 02:30:10    34  2018               1984

Calculate date of birth from age and event date

2 Answers2