Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Saki Tapsell <sakiuchiyama@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Combination of a multinomial logit model and a logit model |
Date | Wed, 3 Oct 2012 11:34:10 +0100 |
Hi, I have two dependent category variables. cc_2 - binary - This is whether a firm has borrowed from friends and family cc_6 - category variable with 6 possible outcomes (credit-constraint statuses), describing whether the firm tried to borrow from the bank, if they did whether or not they succeeded, and if they didn't apply, whether they chose not to for negative reasons (too complicated to apply, or they knew they wouldn't be accepted, etc). As credit constraint status is determined simultaneously, and formal and informal credit market are inter-related, I want to estimate the determinants of credit constraints for formal and informal credit market at the same time. I can separately estimate cc_6 by using multinomial logit model, and cc_2 by a logit model, but I have no idea how to combine the two. Bivariate logit/probit model may be too simplistic. Is there extension of bivariate logit model or some kind of multivariate multinomial logit model in Stata? Thank you very much. Saki * * 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/