Statalist


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

RE: st: clustering with xtabond2


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: clustering with xtabond2
Date   Mon, 12 Oct 2009 14:21:07 +0100

Francesco,

How many clusters to you have?  The asymptotics for the Hansen J-stat
with clustering, liked cluster-robust SEs and other test statistics,
requires the number of clusters to go off to infinity.

If you have very few clusters, you could well get very poor finite
sample performance for your J stat and indeed for all your other
cluster-robust statistics, including your SEs.

HTH,
Mark

Prof. Mark E. Schaffer FRSE
Department of Economics
School of Management & Languages
Heriot-Watt University
Edinburgh EH14 4AS  UK
44-131-451-3494 direct
44-131-451-3296 fax
http://ideas.repec.org/e/psc51.html
 

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Francesco Trivieri
> Sent: Monday, October 12, 2009 11:59 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: clustering with xtabond2
> 
> Dear Maarten,
> 
> thank you for your answer. Of course the p-value I was 
> referring to is 
> greater than 0.1 (and lower tha 0.2).
> 
> Francesco
> 
> 
> ----- Original Message ----- 
> From: "Maarten buis" <maartenbuis@yahoo.co.uk>
> To: <statalist@hsphsun2.harvard.edu>
> Sent: Monday, October 12, 2009 12:33 PM
> Subject: Re: st: clustering with xtabond2
> 
> 
> > --- On Mon, 12/10/09, Francesco Trivieri wrote:
> >> I'm using the xtabond2 command and the Hansen test of
> >> overid. restrictions is not significant, its p-value being
> >> always lower than 0.20 . When I cluster the observations,
> >> yet, the p-value is 1 or very close to it. Is the Hansen not
> >> reliable when  clustering?
> >
> > I don't know the answer, but I do know that small p-values
> > (close to 0) indicate significant results, i.e. a rejection
> > of the null-hypothesis. Large p-values (close to 1) indicate
> > non-significant results, i.e. you are not able to reject
> > the null-hypothesis.
> >
> > (The limit .20 is very unusual, usualy .05 is chose and
> > sometimes .10  or .01)
> >
> > -- Maarten
> >
> > --------------------------
> > Maarten L. Buis
> > Institut fuer Soziologie
> > Universitaet Tuebingen
> > Wilhelmstrasse 36
> > 72074 Tuebingen
> > Germany
> >
> > http://www.maartenbuis.nl
> > --------------------------
> >
> >
> >
> >
> > *
> > *   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/ 
> 
> *
> *   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/
> 


-- 
Heriot-Watt University is a Scottish charity
registered under charity number SC000278.


*
*   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