Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: random effects with clustered s.e.


From   Austin Nichols <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: random effects with clustered s.e.
Date   Mon, 16 Nov 2009 21:15:52 -0500

Shuaizhang Feng <fengsz@yahoo.com>:
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 <fengsz@yahoo.com> 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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index