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 at the end of May, and its replacement, statalist.org is already up and running.


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

st: comparing mixed-effects regression coefficients between models


From   "Eike Mark Rinke" <emrinke@web.de>
To   statalist@hsphsun2.harvard.edu
Subject   st: comparing mixed-effects regression coefficients between models
Date   Tue, 24 Jul 2012 16:24:44 +0200 (CEST)

Dear statalisters,

I would like to compare two coefficients for the same regressor between two multilevel mixed-effects linear models with different outcomes (but otherwise identical specification, and estimated on the same sample, with only minor differences between models due to missing values). However, I do not seem to find a way to accomplish this task, so I am turning to the list for advice.

The situation looks something like this in Stata:

xtmixed y1 a b || id: // model a
xtmixed y2 a b || id: // model b

Is there a way to test for the difference between the coefficients/partial correlations a --> y1 and a --> y2? In other words, is there a way to test if a is significantly more strongly/weakly associated with y1 than with y2? (Using -suest- with subsequent Wald testing does not seem to work in a mixed-effects setting; suest produces the following error upon an attempt to combine estimates for models a and b: >>unable to generate scores for model M19 suest requires that predict allow the score option<<)

I'll be grateful for any thoughts, suggestions, pointers or immediate solutions for this problem.

Eike

-- 
 Ihr WEB.DE Postfach immer dabei: die kostenlose WEB.DE Mail App für iPhone und Android.
https://produkte.web.de/freemail_mobile_startseite/
*
*   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   |   Site index