rdatecharacter2-digit-year

Character 2 digit year conversion to year only

Using R

Got large clinical health data set to play with, but dates are weird

Most problematic is 2digityear/halfyear, as in 98/2, meaning at some point in 1998 after July 1

I have split the column up into 2 character columns, e.g. 98 and 2 but now need to convert the 2 digit year character string into an actual year.

I tried as.Date(data$variable,format="%Y") but not only did I get a conversion to 0098 as the year rather than 1998, I also got todays month and year arbitrarily added (the actual data has no month or day).

as in 0098-06-11

How do I get just 1998 instead?

Solution

  • Not elegant. But using combination of lubridate and as.Date you can get that.

    library(lubridate)
data <- data.frame(variable = c(95, 96, 97,98,99), date=c(1,2,3,4,5))
data$variableUpdated <- year(as.Date(as.character(data$variable), format="%y"))

    and only with base R

    data$variableUpdated <- format(as.Date(as.character(data$variable), format="%y"),"%Y")