Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | sara balestri <sara.balestri@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: ztnb model |
Date | Thu, 27 Oct 2011 18:54:52 +0200 |
Dear Stata users, I'm running a ztnb model on a sample of count data which, by construction, excludes the possibility of zero counts. Overdispersion detected in the data. I have few questions: 1) does exist a nbvargr equivalent command to be applied for nb and ztnb comparison? 2) my count data do not include zero counts and observations start with a minum value of 25 for the dependent variable (by definition of the event analysed) . I'm wondering if a ztnb is the most appropriate model and/or if it is possible to fix the truncation at that value. Any suggestion? thanks Sara Balestri PhD candidate Political Science Faculty Catholic University of Sacred Heart Milan (Italy) * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/