Date:         Mon, 21 May 2001 18:29:26 +0200
Reply-To:     Inke.Boeddeker@med.uni-marburg.de
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Inke Böddeker
              <boeddeke@POST.MED.UNI-MARBURG.DE>
Subject:      Thanks: Multinomial Logit
Comments: To: SAS-L@listserv.uga.edu, allstat@JISCMAIL.AC.UK
Content-type: text/plain; charset=ISO-8859-1

Dear all,
thanks for all your interest and suggestions for my problem on
multinomial logit and probit models.
The most helpful replys were those directing me towards programs
that solve my problem:
Sawtooth Software
Stata using the -constrain- command
COXREG estimation for the conditional logit model
R
Further, I was hinted at the book by Agresti with examples on
discrete-choice models
It seems that (as far as I have tried) Stata actually does the trick. If
anyone is interested in the details, please contact me.
Thanks again for all your help,
Inke
____________________________________________________
Dipl. Psych. Inke R. Böddeker
Institute of Medical Biometry and Epidemiology
Philipps-University Marburg
Bunsenstr. 3, 35037 Marburg
Germany
Tel.: 06421-2865066
Fax: 06421-2868921
mail: boeddeke@mailer.uni-marburg.de