Date:         Tue, 14 Dec 1999 12:19:44 -0800
Reply-To:     Michael Mitchell <mnm@UCLA.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Michael Mitchell <mnm@UCLA.EDU>
Subject:      Y2K and Statistical Computing
Comments: To: statalist@hsphsun2.harvard.edu, spssx-l@uga.cc.uga.edu,
          sas-l@uga.cc.uga.edu
Content-Type: text/plain; charset="us-ascii"; format=flowed

Greetings
   The year 2000 is getting near and we would like to invite you to visit
our Statistical Computing and the Year 2000 pages if you would like more
information about handling dates in the year 2000 and beyond in packages
like Stata, SAS, and SPSS.  We have created web pages to help you assess
whether you have Y2K problems with your research data and, if you do, how
you can solve these problems in SAS, SPSS and Stata.  We invite you to
visit these pages at
http://www.oac.ucla.edu/training/stat/
and then click on "Statistical Computing and the Year 2000".
   In short, those who are at greatest risk are those who 1) analyze data
that contains dates, and 2) the dates are stored using only 2 digits to
represent the year (e.g. 12/25/99 instead of 12/25/1999), and 3) you will
soon encounter dates for the year 2000 (e.g. 1/1/00).  For example, if your
data contains "date of birth", and the
dates are stored in a format like 12/25/99, and you will soon be analyzing
birth records for January 1, 2000 and beyond, then you may soon encounter
Y2K problems in analyzing your data where 1/1/00 will be interpreted as
January 1, 1900.  This could cause results like the age of newborn children
being 100 years old. For further examples of such problems, and suggestions
of how to solve them, please visit our Y2K pages in the link
given above.

Best new year wishes,

Michael Mitchell
UCLA Office of Academic Computing