Re: Fw: st: fixed effects quantile regression

[Maarten Buis <maartenlbuis@gmail.com>]

On Wed, Dec 12, 2012 at 3:52 PM, Lisa Marie Yarnell wrote me privately:
> Hi Maarten,
>
> OK, maybe I should be trying the fixed effects quantile model using "qreg" instead of "sqreg"?  (I have seen this on various websites, so others have done it.)
> If I want estimates for the 50th and 80th quantiles, can I run this using two separate xi: qreg models?  I am guessing that with enough reps, and perhaps repeated runs with different seeds, that reliable estimates for the 50th and 80th quantiles could be determined using two separate runs of the model instead of the simultaneous?

First, the rule on Statalist (see the Statalist FAQ) that you respond
to the list and not privately.

Second, I really don't think that that is the answer, as it does not
address my main point: you want your bootstrap samples to  be samples
of individuals not observations. The fact that other tried this and
got Stata to return numbers to them is no guarantee that those numbers
mean something.

Unless someone else on this list knows another answer, I am afraid
that there is no way around the conclusion that you either need to do
some programming or give up.

Sorry for not giving more pleasant news,
Maarten

---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

http://www.maartenbuis.nl
---------------------------------

*
*   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/