Convert four digit year values to class Date

Question

I've an integer column in my dataset which has four digit year values, like:

 c(2001, 2002, 2002, 2002, 2003, 2005) 

I try to convert the four digit year to class Date using as.Date :

year <- as.Date(as.character(data_file$evtYear), format = "%Y")

But the output is:

"2001-05-15" "2002-05-15" "2002-05-15" "2002-05-15" "2003-05-15" "2005-05-15"

This is giving the wrong output. It's giving two year values in one date (both 2001 and also 15).

I just want the convert my four digit year part from the original data to 'Year' as class Date . Expected out put is simply:

2001 2002 2002 2002 2003 2005 

But their class should be of Date type.

How to achieve this in R?

Answer 1

Based on the comments it turned out that the person asking the question did not need to change a numeric year to "Date" class; nevertheless, the question asked how to do it so here is an answer.

Here are a few ways to create a "Date" class object from a 4 digit numeric year. All use as.Date :

yrs <- c(2001, 2002, 2002, 2002, 2003, 2005)

1) ISOdate

as.Date(ISOdate(yrs, 1, 1))  # beginning of year
as.Date(ISOdate(yrs, 12, 31))  # end of year

This ISOdate solution is a bit tricky because it creates an intermediate POSIXct object so time zone problems could exist. You might prefer one of the following.

2) paste

as.Date(paste(yrs, 1, 1, sep = "-")) # beginning of year
as.Date(paste(yrs, 12, 31, sep = "-")) # end of year

3) zoo::as.yearmon

library(zoo)

as.Date(as.yearmon(yrs)) # beginning of year
as.Date(as.yearmon(yrs) + 11/12, frac = 1) # end of year

Note: If y is the result for any of the above then format(y, "%Y") gives the character year and as.numeric(format(y, "%Y")) gives the numeric year.

Answer 2

As already recognized by the OP, a year alone does not make up a valid date because month and day are not specified.

However, some date and date-time conversion functions, eg, ymd() , parse_date_time() , in the lubridate package recognize a parameter truncated to allow for parsing of incomplete dates:

yrs <- c(2001, 2002, 2002, 2002, 2003, 2005)
lubridate::ymd(yrs, truncated = 2L)

 [1] "2001-01-01" "2002-01-01" "2002-01-01" "2002-01-01" "2003-01-01" "2005-01-01"

The years have been completed by 1st of January to make a valid date. The result is of class Date .

Answer 3

You can do:

library(lubridate)
yrs <- c(2001, 2002, 2002, 2002, 2003, 2005)
yr <- as.Date(as.character(yrs), format = "%Y")
y <- year(yr)

Output:

2001 2002 2002 2002 2003 2005

Answer 4

A lubridate answer:

  library(lubridate)
  year <- ymd(sprintf("%d-01-01",data_file$evtYear))