Bookmark and Share

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

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

Re: st: comparing nested models after multiple imputation

From   Stas Kolenikov <>
Subject   Re: st: comparing nested models after multiple imputation
Date   Sun, 20 May 2012 23:31:39 -0400

On Sun, May 20, 2012 at 3:08 PM, Chelsea Garneau
<> wrote:
> I'm trying to find a way to compare nested models using xtreg, xtlogit,
> xtmelogit, xtmixed after doing multiple imputation.  By "nested models" I
> mean subsequent models with additional predictors added, not multilevel -
> though my data are multilevel data as well. Because the LL's are not
> pooled, there is no e(ll) for the lrtest.  Rubin's rules combine parameter
> estimates and standard errors, not LL's, but what is the best way to test
> nested models using xt commands after mi?

Since MI and Rubin's rule only apply to point estimates and variances,
likelihood ratios are out of question, and you only have Wald tests to
use in this situation. To understand the relations between the
likelihood ratio, Wald, and score/Lagrange multiplier tests, read a
great exposition by Buse from ages ago in The American Statistician:

---- Stas Kolenikov
---- Senior Survey Statistician, Abt SRBI
-- Opinions stated in this email are mine only, and do not reflect the
position of my employer

*   For searches and help try:

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