Statalist


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

Re: st: suppressing constant in qreg


From   "Woolton Lee" <[email protected]>
To   [email protected]
Subject   Re: st: suppressing constant in qreg
Date   Wed, 31 Oct 2007 11:09:46 -0400

On second thought, what I wrote below doesn't make a whole lot of
sense.  What I would like to do is estimate quantile regression
exploiting the fact that I have panel data.  Can anyone tell me if
STATA has add on panel data options available for qreg?

On 10/31/07, Woolton Lee <[email protected]> wrote:
> okay.  Thanks for your reply.  I get the impression that the
> appropriate approach to implementing an analog of "fixed effects" for
> quantile regression would be to remove the respective quantile value
> of the dependent variable.  In other words, if I'm estimating qreg for
> the 0.25 quantile would a more sensible fixed effects analog be to
> subtract the average of the 0.25 quantile of the DV for each panel?
>
> person ID    time      explan. variable  DV
>    1            1998
>    1            1999
>    1            2000
>    1            2001
>    2            1998
>    2            1999
>
> so take the 25th percentiles of the DV in 1998, 1999, 2000, 2001, &
> 2002, sum and divide by 5?
>
> It seems like this could be done to the data first then qreg could be
> run.  What do you think?
>
> Woolton
>
> On 10/31/07, Maarten buis <[email protected]> wrote:
> > --- Woolton Lee <[email protected]> wrote:
> > > Is there any way to suppress the constant term when one estimates
> > > quantile regression using qreg?  I've fixed effects into the equation
> > > and many of the dummy variables are omitted because of collinearity
> > > with the constant term.
> >
> > No, see: http://www.stata.com/statalist/archive/2007-10/msg00809.html
> >
> > Anyhow, even if it were possible I don't think it would be advisable,
> > as the dropping of many dummies suggests more serious problems to me.
> > Also I am not sure if adding dummies will give you a fixed effects
> > model in this case: I always think of a fixed effects model in terms of
> > demeaning, and demeaning doesn't seem quite right in quantile
> > regression (maybe "demedianing").
> >
> > -- Maarten
> >
> >
> > -----------------------------------------
> > Maarten L. Buis
> > Department of Social Research Methodology
> > Vrije Universiteit Amsterdam
> > Boelelaan 1081
> > 1081 HV Amsterdam
> > The Netherlands
> >
> > visiting address:
> > Buitenveldertselaan 3 (Metropolitan), room Z434
> >
> > +31 20 5986715
> >
> > http://home.fsw.vu.nl/m.buis/
> > -----------------------------------------
> >
> >
> >       ___________________________________________________________
> > Want ideas for reducing your carbon footprint? Visit Yahoo! For Good  http://uk.promotions.yahoo.com/forgood/environment.html
> > *
> > *   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index