Hello,
I have been using Kings Relogit to estimate a rare event model for
survey data. I realized that the coefficients produced by the
commands are very similar to the ones obtained using conventional
logit program - only about 0.01 or 0.03 differences between the two.
Which one should I report - I am leaning towards reporting the one
produced by svy: logit command, since that incorporates the full
survey design including stratification information. Is it better to go
this route, or just report the relogit estimates
thanks - cY
------------------------------------------------------------------
*
* 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/
