How about -areg- ?
Something like:
areg y x ,ab(ns) cluster(mst)
This will estimate a fixed effects model and allow observations to be
independent across clusters but not necessarily within them. There is a note
caution in the reference manual dealing with cluster option, see [R]A-F, page
85.
Scott
----- Original Message -----
From: "Edwin Leuven" <[email protected]>
To: <[email protected]>
Sent: Tuesday, March 30, 2004 7:51 AM
Subject: st: fixed effects + clustering
> i want to estimate the following equation
>
> y_{ist} = x_{ist}'b + n_{s} + m_{st} + e_{ist}
>
> with
>
> n_{s} fixed effects
> m_{st} random effects (clusters)
> e_{ist} iid noise
>
> does anyone know how to conveniently estimate this in stata?
>
> xtreg , fe cluster()
>
> would work, but xtreg does not have a cluster option and there are too
> many n_{s}'s to enter them as dummies in a regular regression like
> this
>
> xi: reg y x i.ns, cluster(mst)
>
> thanks, edwin
> *
> * 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/
*
* 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/
