Does anyone have a suggestion for a situation where negative binomial
regression or possibly poisson is appropriate but the data are
sample-selected. For instance, there are a group of respondents sent to a
certain treatment and a researcher is interested in modeling the number of
times those respondents experience event X. Since respondents were sent to
treatment based on some non-random characteristics, there is a need to
account for that selection problem.
A colleague wants to use heckman's correction but I have no experience
mixing that with a negative binomial model. Can anyone weigh in on that?
Much thanks!
David
*
* 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/
