Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: Cluster Option


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Cluster Option
Date   Fri, 5 May 2006 11:09:20 +0100

Max, 

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Max Steinhardt
> Sent: 05 May 2006 09:21
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Cluster Option
> 
> Dear statalist users,
> can someone answer the following question concerning panels:
> is it possible to control for cross-sectional correlation 
> with the cluster option?

"control for" is perhaps not the best phrase to use in connection with the cluster option.  The cluster-robust approach gives you (in certain circumstances) standard errors that are consistent in the presence of heteroskedasticity and (certain types of) serial correlation.  The coefficient estimates are unaffected.

> e.g. my dataset suffers from serial correlation across groups.is using
> cluster(groups) a possible solution for this, or controls 
> cluster only for correlation over time?

I'm not positive I understand properly the problem you're facing, but I suspect this isn't one of the circumstances where this approach works.  Cluster-robust allows for arbitrary heteroskedasticity and within-group correlation, but you still need independence across groups.

Cheers,
Mark

> Best, Max
> 
> 
> --
> GMX Produkte empfehlen und ganz einfach Geld verdienen!
> Satte Provisionen für GMX Partner: http://www.gmx.net/de/go/partner
> *
> *   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/



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