Shuaizhang Feng <[email protected]>:
Perhaps you mean to compare to -xtreg,fe-. In any case, suppose you
use the the GLS random-effects estimator to get coef estimates. Now
you want the SEs, and you can use the conventional or a cluster-robust
version. If you have large numbers of both clusters and obs per
cluster, it probably makes sense to always use the cluster-robust
estimator, but with smaller numbers you might well prefer another
option (perhaps the conventional SEs, perhaps not RE). A simulation
starting with your data is probably the best way for you to answer
this to your own satisfaction.
On Mon, Nov 16, 2009 at 9:04 PM, Shuaizhang Feng <[email protected]> wrote:
> Hi all:
>
> I find that stata 11 allows vce(cluster clustvar) option in xtreg, re models. Has anyone used this before? Does it make sense to cluster s.e. with random effects model? What is the difference with OLS with clustering?
>
> Thank you in advance for any thoughts.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
