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]

Re: st: qreg (median regression)


From   Nick Cox <njcoxstata@gmail.com>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: qreg (median regression)
Date   Wed, 10 Apr 2013 00:01:52 +0100

It so happens that I've been wading through literature on a particular
method in which many papers give simulation results, for a range of
distributions, sample sizes, and so forth. Naturally, that literature
is contradictory and inconclusive. What's best? Depends on your
criteria ... depends on your distribution ... depends on your sample
size.

It strikes me that one advance we really should be exploiting more is,
very simply, not to care too much about that. We can design our
simulations realistic for our problem as easily as -- more easily than
-- those papers. No need to be too deferential about what the
literature says, or even whether literature exists. Fire up your own
simulation!

Or bootstrap.

Nick
njcoxstata@gmail.com

On 9 April 2013 23:46, John Antonakis <John.Antonakis@unil.ch> wrote:
> Thanks Nick.
>
> I also forgot that I a copy of the following:
>
> Hao, L., & Naiman, D. Q. (2007). Quantile regression. Thousand Oaks, Calif.:
> Sage Publications.
>
> This is a good book that explains quantile regression nicely....though I am
> still hunting for the simulations.

On 08.04.2013 00:37, Nick Cox wrote:

>> I think access to literature is made complicated by different names
>> here. Quantile regression,  L_1, median regression are among names
>> used. Much is under the heading robust regression in books on robust
>> statistics.

On 7 April 2013 21:55, John Antonakis <John.Antonakis@unil.ch> wrote:

>>> I have been playing around with Stata's qreg, and have been reading up on
>>> when qreg is useful. For the case of median regression, I can't  find
>>> much
>>> literature on it, particularly about the small sample properties of the
>>> estimator.  Does anyone know of any simulation studies that compares the
>>> efficiency of the median-regression (LAD) estimator to that of the OLS
>>> estimator in small samples (and ideally with data having outliers)?
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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