topic Force a variable as a regressor during stepwise selection in SAS Procedures

Force a variable as a regressor during stepwise selection

wutao9999 — Tue, 31 Mar 2015 21:19:49 GMT

reg data = clean_data

outest = data_a;

model volume = price factor1 factor2 factor3/selection = stepwise;
output out=model_res ;

quit;

I have the above regression model using stepwise selection method. In the model, I want to keep price as a regressor (or explanatory variable) no matter how it is related to volume.

How can I force the price as a regressor as the stepwise step won't remove it? Thank you.

Re: Force a variable as a regressor during stepwise selection

Reeza — Tue, 31 Mar 2015 22:28:41 GMT

Use the INCLUDE option with n=1.

INCLUDE=n

forces the first n independent variables listed in the MODEL statement to be included in all models. The selection methods are performed on the other variables in the MODEL statement. The INCLUDE= option is not available with SELECTION=NONE.

SAS/STAT(R) 9.2 User's Guide, Second Edition