Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: micombine and suest/hausman


From   Ryan Wells <rs_wells@yahoo.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: micombine and suest/hausman
Date   Wed, 1 Oct 2008 08:55:35 -0700 (PDT)

I am using multiple imputation (ice & micombine commands) with data that have a complex design (using svy: commands). I would like to use suest to compare coefficients across models (male versus female). It works when I do it with non-imputed dataset, but does not work when I include micombine before my logit regression command.  here is the syntax and error (again - this error message does not occur with this syntax without micombine):

 
micombine svy,subpop(female):logit expcollege asian hispanic black father_college mother_college byincome ///
parexp par_involv_fac peers_plan_coll siblings ztest_score west midwest south ///
college_prep suburban rural coed magnet privaterel privatenonrel
 
estimates store F
 micombine svy,subpop(male):logit expcollege asian hispanic black father_college mother_college byincome ///
parexp par_involv_fac peers_plan_coll siblings ztest_score west midwest south ///
college_prep suburban rural coed magnet privaterel privatenonrel 
 
estimates store M
 
suest F M
 
 
svy,subpop(female):logit is not supported by suest
r(322);

 
(I've resorted to a convoluted method of putting everything in one model and running a Wald test based on a previous post, but really need to use suest.)


      

*
*   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