Hello,
I am running a multinomial logit on a panel of 1004 observations (49 individuals for roughly 21 years each). I have three categories. In my original regressions, with six right hand side variables, I had no trouble calculating marginal effects and associated standard errors (using the mfx command). I have since added several other variables, and now, although point estimates for the marginal effects are still calculated, no standard deviation can be reported ("fail in calculating second derivative" reads the output). Does anyone know any way around this (i.e., to get a standard deviation for the marginal effect point estimaates)?
By the way, I use Stata 7.0.
Thanks,
Andy Hanssen
--------------------------------------------------------------------
mail2web.com™ - Check your email from the web at http://mail2web.com.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
