Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: BIC for random intercept model


From   Shige <sgsong@spymac.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: BIC for random intercept model
Date   Mon, 31 May 2004 12:34:08 -0700

Dear All,

I have searched the list archieve about how to calculate BIC for regression model with random effects without luck. In cross-sectional case, this can be easily done using "fitstat", but I have not found one single post on how to do it in presence of random effects. In Skrondal and Rabe-Hesketh's (2004, pp.264-265) new book there is some discussion on this, but it's more like "some people did this, other people did that" instead of "we think the best way to do this is". In other words, it is very enlightening for further methodological research but is not extremely helpful for a non-statistician who is working on substantive research question. I would like to have some input from other people on the list. For example, I have a random intercept logistic model with log-likelihood -72717, 32 fixed effects parameter, one random effect parameters, 2059 clusters (aggregate-level units), and 530372 individuals, how to calculate BIC using these numbers?

Any help is greately appreciated!

Best,
Shige Song
Department of Sociology, UCLA
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* 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   |   What's new   |   Site index