Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

st: RE: RE: RE: Residual diagnostics for panel data regression


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: RE: RE: Residual diagnostics for panel data regression
Date   Tue, 25 Sep 2012 10:06:40 +0100

Silvia,

My guess is that testing for heteroskedasticity will involve a fair
amount of work on your part.  The problem is that most of the usual
tests for heteroskedasticity can be seen as special cases of White's
general test, and White's general test is based on the contrast between
the classical VCV and the heteroskedastic-robust VCV.  Maybe this
contrast between a non-robust and a robust VCV has been worked out for
mean-group estimators that -xtmg- reports, but even if it has, it's
probably not been programmed (-findit- didn't reveal anything, at any
rate) and you'd have to roll your own, so to speak.

--Mark

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Silvia Palombi
> Sent: Tuesday, September 25, 2012 8:48 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: RE: RE: Residual diagnostics for panel data regression
> 
> Hi Mark
> 
> Thanks a lot.
> I am using xtcd, but I'm doing xtmg estimation and 
> unfortunately xttest1 etc. can only be used after xtreg and xtgls.
> 
> As an alternative, I constructed a loop to ran -wntestq- 
> (Portmanteau's serial correlation test) on region-specific 
> residuals, stored the p-values and then used these to obtain 
> a Fisher statistics.
> This is as far as I could get because I cannot do the same 
> with -hettest-, since this command can only be used after 
> -reg- and not to test residuals directly as with -wntestq- 
> (i.e., hettest resid is not allowed). 
> 
> If you have any other suggestions it would be helpful to hear them.
> 
> Many thanks,
> Silvia
> 
> 
>  
> ________________________________________
> From: owner-statalist@hsphsun2.harvard.edu 
> [owner-statalist@hsphsun2.harvard.edu] On Behalf Of Schaffer, 
> Mark E [M.E.Schaffer@hw.ac.uk]
> Sent: 24 September 2012 21:34
> To: statalist@hsphsun2.harvard.edu
> Subject: st: RE: Residual diagnostics for panel data regression
> 
> Silvia,
> 
> > -----Original Message-----
> > From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> > statalist@hsphsun2.harvard.edu] On Behalf Of Silvia Palombi
> > Sent: 23 September 2012 22:34
> > To: statalist@hsphsun2.harvard.edu
> > Subject: st: Residual diagnostics for panel data regression
> >
> > Dear Statalisters
> >
> > I would like to run diagnostic tests on residuals from a panel data
> model - in
> > particular normality, serial correlation and heteroskedasticity.
> >
> > For normality, -iqr- and -swilk- seemed to work, but the -wntestq-
> > Portmanteau test for serial correlation and the -hettest- White
> > heteroskedasticity test did not (because the sample 
> includes multiple
> > panels).
> >
> > Could anyone tell me whether there are any post-estimation commands
> > which would allow me to run similar tests on residuals from 
> panel data
> > models?
> 
> A bit of searching with findit yielded some possibilities:
> 
> xttest1
> xttest2
> xttest3
> xtcsd
> xtcd
> xtserial
> 
> -findit [P]- (where [P] is the name of the program above) 
> will tell you
> about the program and how to install it.
> 
> HTH,
> Mark
> 
> >
> > Thank you.
> > *
> > *   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 the Sunday Times
> Scottish University of the Year 2011-2012
> 
> We invite research leaders and ambitious early career researchers to
> join us in leading and driving research in key 
> inter-disciplinary themes.
> Please see www.hw.ac.uk/researchleaders for further 
> information and how
> to apply.
> 
> 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/
> *
> *   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 the Sunday Times
Scottish University of the Year 2011-2012

We invite research leaders and ambitious early career researchers to 
join us in leading and driving research in key inter-disciplinary themes. 
Please see www.hw.ac.uk/researchleaders for further information and how
to apply.

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   |   Site index