Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

st: SUR for GLM, but not SUEST


From   Mohamad Mahmoud <ksgksg@gmail.com>
To   statalist <statalist@hsphsun2.harvard.edu>
Subject   st: SUR for GLM, but not SUEST
Date   Wed, 9 Apr 2014 13:03:53 -0400

Hi,
I am trying to estimate two systems of equations on the effect of
holding a certain occupation on turning out to vote then the effect of
holding that occupation on voting for a certain candidate .
Since the dependent vars (voting and voter turnout turnout) are
proportions per district, I am using -glm- with logit link and
binomial family.
I'd like to use seemingly unrelated regression (SUR) to take into
account the correlation across equations since there is a correlation
between the effect of that occupation on turning out to vote and who
one voted for. However, given that I am using GLM, I cannot use
-sureg-.
I looked into -suest-, but it is giving me the exact results from
estimating the equation simultaneously. I somewhat understand why form
this stata post:
http://www.stata.com/statalist/archive/2007-08/msg00245.html

I am not concerned about efficiency since my results are highly
significant nor do I want to conduct cross equation tests, but I want
to make sure my approach is correct and my results are robust.
So my questions are:
1) Is SUR the best approach?
2) If so, how can I do -sureg- for a glm estimation in Stata (-suest-
won't work since it does not take cross equation correlation).

thank you,
Mohamad Mahmoud
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index