| Date: | Wed, 14 May 2003 13:42:50 -0500 |
| Reply-To: | Rodney Sparapani <rsparapa@MCW.EDU> |
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
| From: | Rodney Sparapani <rsparapa@MCW.EDU> |
| Organization: | Medical College of Wisconsin, Milwaukee |
| Subject: | Re: Dates from the first Century |
| Content-Type: | text/plain; charset=us-ascii; format=flowed |
Ed Heaton wrote:
>Okay, why doesn't
>
> x = input( "03150044" , mmddyy8. ) ;
>
>set X to missing?
>
> x = input( "01010100" , mmddyy8. ) ;
>
>sets X to missing and gives a note that the argument is invalid.
>This I expect because SAS has no date before 1582. This I expect.
>But why don't I get a similar result from
>
> x = input( "12310099" , mmddyy8. ) ;
>
>I know why not ... but why not?
>
>Ed
>
>
Because the YEARCUTOFF option applies to two-digit years so you can
never have 1st century dates. I suspect that data processors back then
were struggling with the Y1C problem ;o) Seriously, if you need such early
dates you can't use SAS dates because there is little agreement on
"dates" before 1582 as you noticed. There is another method based
on the Julian period which starts at the beginning of recorded history.
--
Rodney Sparapani Medical College of Wisconsin
Sr. Biostatistician Patient Care & Outcomes Research
rsparapa@mcw.edu http://www.mcw.edu/pcor
Was 'Name That Tune' rigged? WWLD -- What Would Lombardi Do
|