I would really urge you to use the recently-developped xtivreg2 for
this purpose. It will generate heteroskedasticity and auto-correlation
robust se's. Just ommit the IV options...it will estimate fixed effect
specifications, while providing you with standard errors you (and
others) can actually believe. You'll have the choice between
heteroskedasticity/AC of unknown form through the robust cluster()
option, or if you are willing to put a bit more structure on the
auto-correlation patterns in the residuals, you'll have a bunch of
HAC-VCV options open...with xtivreg2, microeconometricians simply have
no excuse anymore for flawed inference, at least in linear models.
Regards,
Pierre
-----------------------------------------------------------------------------------------------------------
Pierre Azoulay
Associate Professor of Management
Columbia University Phone: (212) 854-9684
Graduate School of Business Fax: Don't send any
3022 Broadway, Uris Hall 704
New York, NY 10023 http://www.columbia.edu/~pa2009/
Hi statalistuser,
i have a general question concerning panel estimations: when i use xtreg are
there other possibilities to control the standard errors for
heteroskedasticity/autocorrelation than using robust se if I do not want to
use instrumental variables?
Thanks, Michael
*
* 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/
