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]
Re: st: oglm and heterogeneous choice models
From
Richard Williams <[email protected]>
To
[email protected], [email protected]
Subject
Re: st: oglm and heterogeneous choice models
Date
Tue, 25 Oct 2011 13:51:09 -0500
At 12:15 PM 10/25/2011, Rourke O'Brien wrote:
Thanks so much for the follow up!
Yes the DV is dichotomous. Does the Brant test work for dichotomous
variables? I kept getting errors that there had to be multiple levels
(ordinal).
The brant test doesn't make any sense with a dichotomous dv; there is
nothing to test when there are only 2 possible values. Basically
there is only one line/regression so there are no parallel
lines/regressions to look at.
For that matter there is no particular reason to use gologit2 with a
dichotomous dv either, unless maybe you want to take advantage of
some of the few things it can do that logit can't. If you look at
your gologit2 results you should see they are the same as your
logit/ologit results; if not let me know.
If I use gologit2 to identify problematic variables can I also use it
to estimate a model that corrects for the hetero (by denoting which
vars meet the pl assumption and which do not)? Or do I need to use
that information and run the model in oglm?
If the dv is a dichotomy, I don't think gologit2 can help you
identify problem vars. With a dichotomous dv, any gologit2 messages
you get about a var not meeting the parallel lines assumption may be
deceptive, since there is really nothing to test.
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam
*
* 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/