(I previously sent this message from my univ account but incoming email is problematic now. Sorry.)
I have the following probit model:
use of violence by actor in a crisis (1,0)= duration of actors involvement in the crisis, contiguity, domestic unrest, democracy, etc, cluster(crisis) robust
observations (n=712) are actors in intl crises, 1918-1994. duration is highly sig and positive in this probit.
i then seek the determinants of duration. i specify violence on the right side of a gamma regression. i test
streg violence, democracy, domestic unrest, etc, cluster(crisis) robust
violence has a positive impact on duration.
so i end up with duration increasing chance of violence and violence increasing duration.
how can i sort this out? seems like standard tests of endogeneity like smith-blundell wont work because of the duration variable.
thanks in advance,
Karl
[email protected]
____________________________________________________________
Get advanced SPAM filtering on Webmail or POP Mail ... Get Lycos Mail!
http://login.mail.lycos.com/r/referral?aid=27005
*
* 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/
