st: Re: RE: Re: Quantile regression for panel data

["Jeannette Wicks-Lim" <janetlim@econs.umass.edu>]

I'm not sure about the theory underlying how the sampling variance is 
estimated with qreg, but as far as I know, the only way to deal with 
clustering in your data within the context of quantile regression, is to 
bootstrap. Besides the bs command you can also look into clad (Censored 
Least Absolute Deviations estimator with bootstrapped se).
J
----- Original Message ----- 
From: "FEIVESON, ALAN H. (AL) (JSC-SK) (NASA)" <alan.h.feiveson@nasa.gov>
To: <statalist@hsphsun2.harvard.edu>
Sent: Tuesday, May 17, 2005 2:48 PM
Subject: st: RE: Re: Quantile regression for panel data



> Unfortunately (at least in Stata 8) qreg does not allow a -cluster- 
> option.
> I am wondering why - is there some theoretical reason for this or is it
> simply a matter of nobody having provided the code for it in -qreg- ?
>
> Al Feiveson
>
> . qreg y delph,cluster(jsub)
> Iteration  1:  WLS sum of weighted deviations =  5381.2885
>
> option cluster() not allowed
> r(198);
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu]On Behalf Of Jeannette
> Wicks-Lim
> Sent: Tuesday, May 17, 2005 1:36 PM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Re: Quantile regression for panel data
>
>
> One way to deal with clustering, I believe, is to bootstrap your standard
> errors using the cluster option (for info, just type "help bs").
>
> Jeannette
> ----- Original Message ----- 
> From: <stefanie.brodmann@upf.edu>
> To: <statalist@hsphsun2.harvard.edu>
> Sent: Tuesday, May 17, 2005 2:28 PM
> Subject: st: Quantile regression for panel data
>
>
>
> > Dear All,
> >
> > for my thesis, I would like to estimate the returns to education for
> > immigrants
> > and natives at different points in the earnings distribution. I have a
> > large
> > panel data set with 85000 observations. I want to make use of quantile
> > regression techniques; however, I am not sure if this works with
> > longitudinal data.
> >
> > Is anyone of you aware of panel data features for quantile regression? Or
> > is
> > there anything like the cluster-option for quantile regression in order 
> > to
> > take
> > account of the dependence among individuals?
> >
> > I am very grateful for any comments. Thanks a lot!
> >
> > Stefanie
> >
> >
> >
> >
> >
> > *
> > *   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/
> *
> *   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/