Glenn--
Try
xtreg ......, i(id) fe cluster(id)
ssc inst ivreg2
ssc inst xtivreg2
xtivreg2 ......, i(id) fe cluster(id)
On 5/3/07, Schaffer, Mark E <[email protected]> wrote:
First, Stock and Watson have shown that the usual heteroskedastic-robust
SEs for the fixed effects estimator are not consistent, but the
cluster-robust ones are. See
http://www.nber.org/papers/T0323
On 5/22/07, Glenn Hoetker <[email protected]> wrote:
I have unbalanced panel data and thus need to control for inflated
standard errors. Three means of doing so suggest themselves:
xtreg ......, i(id) fe
xtreg ......, i(id) re
xtreg ......., cluster(id)
The discussion under [R] regress demonstrates all three and indicates
that they are alternative means (unless the assumptions of random
effects are rejected by a Hausman test, which they are in my case).
However, it does not discuss what would lead one to prefer one over
the other--an insight I've not found anywhere else either.
*
* 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/
