if you want the best possible fit, use logit. relogit gives up some of the
fit in order to get unbiased estimates. this is like weighted least
squares vs least squares, the latter fitting as well as possible, the
former getting you a more efficient estimator.
Gary
--
*Gary King* - Albert J. Weatherhead III University Professor - Director, IQSS
<http://iq.harvard.edu>- Harvard University
GKing.Harvard.edu <http://gking.harvard.edu/> - King(a)Harvard.edu -
@kinggary<http://twitter.com/kinggary>- 617-500-7570 - Asst 495-9271 -
Fax 812-8581
On Tue, May 15, 2012 at 4:09 PM, <partovilatif(a)ut.ac.ir> wrote:
Dear all,
I am a PhD student in Tehran University who is currently working on my
dissertation. I have estimated a Relogit regression of the determinants of
occurrence of injuries by migrants and non-migrants. In my work I try to
compare “goodness of fit” in regression models for two groups. I
would like to know:
1) If there is a way to test the “goodness of fit” for regression
model by migrants and non-migrants with Relogit? If yes, How to get the
“goodness of fit” after running Relogit by migrants and non-migrants?
2) Is it possible to control “unobserved heterogeneity”, in another
word, to control for unobserved characteristics due to omitted variables
with Relogit? If yes, what do I do?
Many thanks,
Latif Partovi
-
---
relogit mailing list served by HUIT
List Address: relogit(a)lists.gking.harvard.edu
Subscribe/Unsubscribe:
http://lists.gking.harvard.edu/mailman/listinfo/relogit
Relogit mailing list
Relogit(a)lists.gking.harvard.edu
https://lists.gking.harvard.edu/mailman/listinfo/relogit