Date:         Fri, 23 Oct 2009 01:14:08 -0700
Reply-To:     shiva <shiva.saidala@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         shiva <shiva.saidala@GMAIL.COM>
Organization: http://groups.google.com
Subject:      Re: MACRO FOR ALL POSSIBLE REGRESSIONS XXXX
Comments: To: sas-l@uga.edu
Content-Type: text/plain; charset=ISO-8859-1

Hi Dan,

How about using stepwise regression.

proc reg data= final;
model salesqty = var1 var2 var3 var4/selection=stepwise;/*all the
variable here are significant variable of the regression model*/
output out=test p=A1res;
run;


Thanks,
shiva