|Date: ||Mon, 17 Dec 2007 15:42:08 -0600|
|Reply-To: ||"Paul A. Thompson" <paul@WUBIOS.WUSTL.EDU>|
|Sender: ||"SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>|
|From: ||"Paul A. Thompson" <paul@WUBIOS.WUSTL.EDU>|
|Subject: ||Re: Performing thousands of tests automatically|
|Content-Type: ||text/plain; charset="us-ascii"|
While Mary is correct in that you could do this with a macro, the COST of
doing such would be unbelievably high over the BY-Variable approach. That's
because there are 100,000 dataset access times to be added in. With
BY-Variables, you open the dataset 1 time, and run multiple times WITHIN the
procedure. With the MACRO approach, you have 100,000 extra dataset open
times to be added in.
With 10 - 20 - 30 regressions, no worries, mate. With 100,000 it could be a
substantial SUBSTANTIAL addition of run time.
Paul A. Thompson, Ph.D.
Division of Biostatistics, Washington University School of Medicine
660 S. Euclid, St. Louis, MO 63110-1093
From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Ron Do
Sent: Monday, December 17, 2007 3:40 PM
Subject: Re: Performing thousands of tests automatically
Thanks everyone for the responses. I will try some of the suggestions and
see how they work.
Mary is right, this is for a genome-wide association study using over
100,000 markers. There are some genetics programs that can handle this
data, but I'm interested in applying a mixed effects model to analyse this
data and hence am curious if SAS is capable of doing this.
I'll let you all know how it goes.