st: Comparing different random effect xtnbreg models

[Alberto La Rosa <alberto.larosa311@gmail.com>]

Goodmorning to all,

I'm a Stata beginner and I'm using version 12. I've read statalist FAQs
and I hope you could help me even if my questions seem elementary.

I performed some random effect xtnbreg regressions on panel data and I
should choose the better one.

For example if I have 3 models

**********example begin*********
xtnbreg depvar i.firm i.yearx1 x2 x3, re

xtnbreg depvar i.firm i.year x1 x3 x4, re

xtnbreg depvar i.firm i.year x2 x4 x5, re

**********exampleend*********

So, my questions  are:
1) how can I compare the models in order to choose the better one? (I've 
made a search in previous threads but I've still doubts..is it 
sufficient comparing models BICs?)	
2) is it a problem if only the dummy variables (only firm and year) have 
not significative beta?

Thanks a lot for your time and consideration

Albert
*
*   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/