I think, if you specify the robust option, like gllamm, robust you will
get robust standard errors. Or you can also use clustering option with
your gllamm. If I got your question!
mayten
I think it's safer to assume fixed effects and restrict your inferences to
> those 10 countries. The countries in the EU differ so much so that if you
> had a different sample, you may end up with very different estimates.
>
> Tim
>
>
>
>
> martina brandt <[email protected]>
> Sent by: [email protected]
> 21/12/2006 16:16
> Please respond to
> [email protected]
>
>
> To
> [email protected]
> cc
>
> Subject
> st: gllamm and standard errors
>
>
>
>
>
>
> Hello!
> In a multilevel-regression we have a sample of 10 countries (observations
> on the highest level), the total population is the European Union - so
> we have information about almost half of the total population. Std. Errors
> should thus be corrected for the fact that the size of the total
> population is quite small. Does anyone know, how to calculate those
> corrected SE using gllamm?
>
> Thanks in advance!
>
> Klaus & Martina
>
> *
> * 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/
>
*
* 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/
