Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Similar Relogit vrs Logit Estimates: Which to Report?


From   Chao Yawo <Yawo1964@yahoo.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Similar Relogit vrs Logit Estimates: Which to Report?
Date   Thu, 16 Jul 2009 13:59:42 -0400

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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index