Relative risks for multinomial logit regression

[Demographer]

Hello,

I know it is possible to estimate relative risks from a binomial logistic regression using proc genmod, as explained there:

http://www.ats.ucla.edu/stat/sas/faq/relative_risk.htm

Is it also possible to do something similar for a multinomial logit regression?

Thank you

[StatDave · Accepted solution]

Yes.  See this note.  The answers to many questions can be found in the Samples and SAS Notes in our searchable knowledgebase.  You can use the search engine there to find the answers you need.  See also our list of Frequently Asked for Statistics, FASTats.