[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Using Bayesian Information Criterion (BIC)

From   Maarten buis <>
Subject   RE: st: Using Bayesian Information Criterion (BIC)
Date   Tue, 22 Apr 2008 18:12:13 +0100 (BST)

--- Lorena Barberia <> wrote:
> I would like to use the BIC to compare different estimations.
> Specifically, I am trying to estimate models with panel corrected
> standard errors (XTPCSE) and GMM First Difference and System
> estimators. After both commands, I tried to issue the command estat
> ic. I receive the following error message "likelihood information not
> found in last estimation results." 

The BIC is a function of the log likelihood and the number of
observations, so in order to compute the BIC you need the
log-likelihood and the number of observations. The log likelihood is
naturally available when you estimate a model with maximum likelihood,
and sometimes you can derive the log-likelihood when another estimation
method has been used, for instance -regress- uses Ordinary Least
Squares but you can still derive the log-likelihood. However, this is
not true for all other methods, and apperently you ran into such a
case, or maybe it is possible but it has simply not been implemented.

Hope this helps,

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Sent from Yahoo! Mail.
A Smarter Email
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index