Date:         Fri, 13 Sep 2002 14:25:39 -0700
Reply-To:     Cassell.David@EPAMAIL.EPA.GOV
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "David L. Cassell" <Cassell.David@EPAMAIL.EPA.GOV>
Subject:      Re: missing values
Content-type: text/plain; charset=us-ascii

Hollie Rich <hx97403@HOTMAIL.COM> wrote:
> I have a huge data set (about 1300 variables and 712 observations). I
need
> to replace the all missing vaules in data set with -9. I wonder is
there
> easy way to do?

Jack already told you how, and said what I was thinking about
data set size.  (The meaning of 'huge' for data sets increases
exponentially with time.)

But I wanted to add something else.  Must you change those
missing values to -9 ?  This is *not* a good thing.  Even
if you're just having to dump results out for Excel users,
who can't handle anywhere near 1300 columns.  Making sure
no one ever accidentally includes osme of those -9 values in
an analysis is impossible.  Making sure no one uses a SAS
missing value is a convenient default.  If you're doing something
for which SAS missing values cause problems, there ought to be
a better approach than this.  I've had to "un-correct" this sort
of fake missing value way too many times.

HTH,
David
--
David Cassell, CSC
Cassell.David@epa.gov
Senior computing specialist
mathematical statistician