Date:         Tue, 1 Feb 2000 07:23:45 GMT
Reply-To:     Victor Aina <aina@SFU.CA>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Victor Aina <aina@SFU.CA>
Organization: Simon Fraser University
Subject:      Help: Computing standard deviation and means

Hello everyone:

I have 20 variables (say x1 x2 x3 ... x20) and 80,000
(eighty thousand) observations.
Now, I need to compute the mean and standard deviation
of each observation. In other words, I want to compute
these statistics for each row. NOT for each variable--
PROC MEANS would do that.

I've written a code in PROC IML that works fine but
for only 10 variables. When I include  anything more
than 10 variables SAS chokes; complaining of lack of
sufficient memory.

Is there a simple way to avoid PROC IML and its huge
demand for memory and yet get the mean and std deviation
computed for each of the 80,000 observations?

Thanks for all pointers.

Victor.

--


+----------------------------------------------------------+
| victor aina  |  e-mail: aina@sfu.ca | fax:(604) 291-5944 |