CODEWITHSUNDEEP
rubuntu
user3083324
user3083324

Reputation: 585

convert date into numbers

I have a bunch of dates:

dates
 [1] "24/06/2010" NA           NA           NA           "11/05/2005" NA           "24/02/2006"
 [8] NA           NA           NA           "27/10/2008" "09/05/2009" NA           NA          
[15] "14/12/2008" "14/12/2008" NA           "07/04/2009" NA

And wanna make a new vector where the dates are transformed into numbers and the NA transformed into -1 values. I know that in Excel dates are converted into numbers according to a standard procedure, here I don't know.

So for example if we fix as starting date 01/01/2008 as number 0 and count from there

24/06/2010 would be 905 days from the starting date.

Upvotes: 0

Views: 115

Answers (2)

Jörg Mäder
Jörg Mäder

Reputation: 677

[Updated] Does this fit your question?

ifelse(is.na(dates),-1,as.Date(strptime(dates,'%d/%m/%Y'))-as.Date("2008-01-01"))

Upvotes: 0

Sven Hohenstein
Sven Hohenstein

Reputation: 81693

Here's an approach:

# the reference date
starting_date <- as.Date("2008-01-01") 
# transform the character vector to a date vector and subtract the reference date
res <- as.Date(dates, "%d/%m/%Y") - starting_date
# replace NAs with -1
res[is.na(res)] <- -1

The result:

# Time differences in days
#  [1]  905   -1   -1   -1 -965   -1 -676   -1   -1   -1  300  494   -1   -1  348
# [16]  348   -1  462   -1

Upvotes: 3

Related Questions