Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

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

Re: st: non-proportional odds and multilevel models

From   "JVerkuilen (Gmail)" <>
Subject   Re: st: non-proportional odds and multilevel models
Date   Tue, 13 Nov 2012 08:02:06 -0500

On Tue, Nov 13, 2012 at 2:52 AM,  <> wrote:

> What options do I have?

I guess it depends. I've often found that that test is rejected but
the nature of the rejection is fairly trivial. It might be a sign that
you have an omitted variable. You may want to lean on your model
specification some more. You could also need a different link such as
cloglog instead of logit. As another poster said, the gologit approach
could be used.

I think it's important to check whether the model is good before
switching to a different link function or non-proportional odds.
*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index