Date:         Sun, 14 Mar 2004 14:47:25 -0500
Reply-To:     Nageswara Rao Punnani <npunnani@MEMPHIS.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Nageswara Rao Punnani <npunnani@MEMPHIS.EDU>
Subject:      proc regression macro ..

hi all ,

can someone please help me out in solving the following problem ..i am
trying to write a macro for this proc reg step ..the macro should be able
to take in  the variables as parameters  and generate the combinations of
all the models possible ..the macro should generate 2 to the power of n
minus 1 no. of models .
eg. %xyz(x1,x2,x3,x4) macro should generate 15 model statements inside the
proc reg statement .
the following proc step actually takes in 3 variables and generates 7
statements ..

proc reg data=z;
       M1: model y=x1;
    M2: model y=x2;
    M3: model y=x3;
    M4: model y=x1 x2;
    M5: model y=x1 x3;
    M6: model y=x2 x3;
    M7: model y=x1 x2 x3;
                             run;
i hope i was able to put across my problem properly ..

i would really appreciate if someone can help me out in this

thanks
rao.