Hello
Iwould be greatfull if someone helps me with the following problem:
I'm running a hausman test to choose between a random effects and fixede
ffects . My sample incluse 141000 observation . I read tht hausman test is
invalid if there is any hetero and serielle correlation . So i used areg
robust to get a robust standard errors for fixed effect .After that i used
Xtrge pa robust (which is equivalent to Xtgee ) to get a robust standard
errors for random effects. Finally i run a hausman test and it gives me
chi 2 = 1 .could xtgee be a substitute for a random effect when we want
to get a robust standard errors ? Note that xtreg re does not support
robust option .
is my procedure correct?
1) areg , robust
2) estimates store fixe
3) Xtgee ( or xtreg,pa robus
4) Hausman
Thanks for help
Hanaa
MSc Finance
*
* 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/
