Dear all,
I use STATA 9.0. I would like to know if there is a way to generate
robust standard error due to clustering in XTIVREG. I know
that XTIVREG will generate the correct standard error assuming that errors
are correlated within group. But the error terms in my case are also not
independent across a broader category (example, group is county but
standard errors are also correlated within state level).
Previously, I always used AREG instead of XTREG when I need to estimate a
fixed-effects model and the clustering is at a higher level. But AREG
cannot handle IV estimation. I used XTIVREG, which does not allow for
cluster option, so I am getting standard errors that are too small.
Thank you very much for the help.
Yu-Chu
*********************************
Yu-Chu Shen
(831) 656-2951
Faculty Research Fellow
NBER/Health Care
Assistant Professor of Economics
Naval Postgraduate School
Monterey, CA 93943
*********************************
*
* 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/
