Re: st: some questions about gllamm

[Dan MacNulty <macn0007@umn.edu>]

Sami - you can perform a likelihood ratio test with gllamm or xtmixed as 
follows:

(1) run the full model and store its parameter estimates:

estimates store model1

(2) run the reduced model and stores its estimates:

estimates store model2

(3) run the likelihood ratio test, assuming that model2 is nested within 
model1:

lrtest model2 model1

good luck,
Dan MacNulty
University of Minnesota





stata_user stata_user wrote:

> Dear All,
> I am using for the moment the gllamm model to fit a random intercept 
> model with individual nested within countries
>
> Dependent variable is continuous
>
> independent variables  are continuous and discrete, I would like to 
> compare between different models (adding more covariates ect....)
>
> How I can use log likelihood test?
>
> My second question concerns checking collinearity? the vif command is 
> not working for the gllamm, how to test if there is collinearity among 
> the covariates?
>
> Thanks a lot for any suggestion
>
> Sami
>
> _________________________________________________________________
> Windows Live™ Messenger has arrived. Click here to download it for free! 
> http://imagine-msn.com/messenger/launch80/?locale=en-gb
>
> *
> *   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/
*
*   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/