Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Brent Gibbons <sephjoe@gmail.com> |
| To | statalist@hsphsun2.harvard.edu |
| Subject | st: Question About Margeff with nbreg |
| Date | Sun, 20 Mar 2011 15:57:50 -0400 |
I ran an nbreg, using the mean dispersion option, with 2 r.h.s. variables plus the intercept (no offset variable). I got highly significant coefficient estimates, but the p-values for the corresponding margeff estimates (of ave. marginal effects over the sample) were MUCH higher and the confidence intervals for these margeff estimates were very large. What might cause this disparity between the results for the coefficient estimates and the results for the margeff estimates? Brent Gibbons * * 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/