Dear Statalist members,
I have to model a categorical dependent variable that has 4 potential outcomes, but those outcomes are not mutually exclusive meaning that an observation may be connected to outcome 2 and 3 or 1 and 4. I thought about using a sequential logit but the possible outcomes do not necessarily follow a sequential pattern. Alternatively I also considered using several separate logit models, but I believe it is not the best alternative either. Would someone have a suggestion?
Best,
Alexandre
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
