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]

Re: st: xtmelogit: comparing models


From   Luca Campanelli <l.campanelli@yahoo.it>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: xtmelogit: comparing models
Date   Sat, 6 Oct 2012 15:56:53 +0100 (BST)

Joerg, Stas, thank you very much for your helpful answers. 

Joerg, yes, I looked at the observed proportions and I made a graph. An interaction groupXcondition makes sense theoretically (that’s my prediction) and it makes sense seen the descriptive stats and the graph. 
Also, thank you for your piece of advice and info about -margins-. It’s very helpful. 

Stas, thank you for the clarification about Wald test vs. Likelihood Ratio test. It’s also important for me to know that, in this case, AIC and BIC should not be taken too seriously. 
I just run the Wald test (using -test-) and it’s what I will report, as it seems it’s more appropriate. In any case, -lrtest- and -test- produced consistent results, which is good. 
At the end this is the command that I run: 
test  1.group#2.condition  1.group#3.condition


Again, thank you very much for your help, 
Luca

*
*   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 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index