Programming the statistical procedures from SAS

PROC CATMOD: unclear parameter names (B1, B2...)

Reply
Contributor
Posts: 52

PROC CATMOD: unclear parameter names (B1, B2...)

Hi there,

I used PROC CATMOD to run a model. The parameter names from the OUTEST are procuded as B1, B2, B3, B4, B5, B199, B200, B201. Don't know why this procedure produces the parameter estimates in these names. It is very confusing for people to identify which is which. Is there any way to use some option to improve this? Thanks in advance.

PROC CATMOD DATA=city;

RESPONSE/OUTEST=work.params;

MODEL group=sex married income region/NOITER NORESPONSE NODESIGN NOPROFILE;

RUN;

QUIT;

SAS Employee
Posts: 241

PROC CATMOD: unclear parameter names (B1, B2...)

If you are fitting a logistic model to a categorical response, I strongly recommend that you use PROC LOGISTIC rather than PROC CATMOD regardless of whether your response has two levels or several levels that are ordered (ordinal) or unordered (nominal). LOGISTIC can fit the same model as CATMOD, but LOGISTIC is more efficient, easier to use (especially when you want predicted values), has more options specialized for logistic models such as effect selection (forward, backward, stepwise, all subsets), and produces odds ratios and ROC curves. If your response has several levels (is multinomial), PROC LOGISTIC fits the ordinal model by default. If it is unordered, use the LINK=GLOGIT option in the MODEL statement to fit a generalized logit model as in PROC CATMOD. See the discussion and examples in the

LOGISTIC documentation:

   SAS 9.2:
   http://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/logistic_toc.htm

   SAS/STAT 9.22 (beginning in SAS 9.2 TS2M3):
   http://support.sas.com/documentation/cdl/en/statug/63347/HTML/default/logistic_toc.htm

   SAS 9.3:
   http://support.sas.com/documentation/cdl/en/statug/63962/HTML/default/logistic_toc.htm

The OUTEST= data set produced by PROC LOGISTIC has parameter names using the variable and level names.

Ask a Question
Discussion stats
  • 1 reply
  • 116 views
  • 0 likes
  • 2 in conversation