home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!IRISHMVS.BITNET!GMMM5K
- Message-ID: <SAS-L%92090915025521@UGA.CC.UGA.EDU>
- Newsgroups: bit.listserv.sas-l
- Date: Wed, 9 Sep 1992 14:02:00 EST
- Reply-To: GMMM5K@IRISHMVS.BITNET
- Sender: "SAS(r) Discussion" <SAS-L@UGA.BITNET>
- From: GMMM5K@IRISHMVS.BITNET
- Subject: logit models
- Lines: 12
-
- To Whom it may concern
- Assuming that one needs to use a logitistic regression model to
- estimate data with a qualitative dependent variable (a variable
- that is coded 0 or 1), which SAS procedure should be used when
- half of the independent variables are dummies (coded 0 or 1) and the
- other half are continuous variables? Based on the SAS/STAT Vol 1
- manual, p 36, it would seem that in reading the information on the
- various models PROBIT, LOGISTIC and CATMOD, that PROBIT and CATMOD
- are the only procedures that allow this? Is there anyway to transform
- the data so that LOGISTIC can be run instead?
- Thank you for your attention and your answers.
- GMMM5K@IRISHMVS.CC.ND.EDU
-