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 at the end of May, and its replacement, statalist.org is already up and running.


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

RE: RE: st: Beta values in QREG


From   "Santos Silva, J.M.C." <jmcss@essex.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: RE: st: Beta values in QREG
Date   Mon, 24 Jun 2013 21:50:46 +0100

Dear Scott,

Indeed it looks like you have a bug in your code that is responsible for the
strange results. As for Maarten's argument of "clash of logics", I think I do
not agree with it for two reasons. First, you may be interested in estimating
quantile regression even if the data have finite second (and higher order)
moments; Stata's manual is a somewhat misleading in this respect but
quantile regression is much more than just a robust "alternative" to mean
regression. Second, even if you are using quantile regression because
your y does not have finite second moments, that does not mean that your 
regressors also do not have second moments; so you may still be able to 
standardize the Xs even if the y has no moments. Having said that, only you 
can know whether it makes sense to standardize the particular regressors 
you have in your model.

All the best,

Joao



> From           "Scott Holupka" <scott.holupka@jhu.edu>
> To           <statalist@hsphsun2.harvard.edu>
> Subject           RE: st: Beta values in QREG
> Date           Mon, 24 Jun 2013 09:30:37 -0400
> 
> No, I'm not getting the same results, which is what I expected, so I wanted
> to be sure there wasn't a more fundamental problem I was missing or if I had
> just made some mistake in my coding.
> 
> Based on your response, as well as Maarten's, I think my answer is both
> that: a) I made some sort of coding mistake that's throwing off the results,
> and b) even if I compute it correctly it may not make sense to use because
> standardized scores since the reason for using quantile/median regression is
> that there is evidence of non-normality, so why adjust by a measure of
> central tendency that will be biased.
> 
> Thanks for the advice.
> 
> Scott
> 
> 
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Santos Silva,
> J.M.C.
> Sent: Sunday, June 23, 2013 5:09 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: RE: st: Beta values in QREG
> 
> Dear Scott,
> 
> What exactly do you mean when you say that running qreg with standardized
> variables leads to different results? The results should be exactly the same
> (I mean, the R2, objective function, etc), except that the coefficients of
> the regressors are now multiplied by the standard errors of the original
> regressors.
> Don't you get this?
> 
> All the best,
> 
> Joao
> 
> 
>> From           "Scott Holupka" <scott.holupka@jhu.edu>
>> To           <statalist@hsphsun2.harvard.edu>
>> Subject           st: Beta values in QREG
>> Date           Fri, 21 Jun 2013 16:03:55 -0400
>> 
>> Does anyone know if there is a statistical reason why the Stata
>> quantile regression program "qreg" does not provide an option for
>> producing beta values?  I know a question about beta values in qreg
>> was raised just a few months ago, and the one response suggested that
>> there might be a statistical reason why the option wasn't available,
>> but I didn't see anything more definitive.
>> 
>> I did try standardizing all of my variables and re-running QREG, as
>> had been previously suggested, but the results between the
>> unstandardized and standardized models seem so different I'm not sure
>> if I did something wrong or if there's a more fundamental reason why the
> results don't line up.
> 
>> Thanks for any advice.
>> 
>> Scott Holupka
>> 
>> 
>> C. Scott Holupka, Ph.D.
>> Senior Research Associate
>> Johns Hopkins University
>> Institute for Policy Studies
>> 3400 N. Charles St.
>> Baltimore, MD 21218-2688
>> 
>> 410-516-5046
>> Scott.Holupka@jhu.edu
>> 
>> 
>> *
>> *   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/
>> 
>> Follow-Ups:
>> Re: st: Beta values in QREG
>> From: Maarten Buis <maartenlbuis@gmail.com>
>> 
>> Prev by Date: Re: st: RE: count number of working adult Next by Date:
>> RE: st: RE: count number of working adult Previous by thread: st:
>> count number of working adult Next by thread: Re: st: Beta values in
>> QREG
>> Index(es):
>> Date
>> Thread
>> 
> 
> *
> *   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/
> 
> *
> *   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/
> 
> References:
> RE: st: Beta values in QREG
> From: "Santos Silva, J.M.C." <jmcss@essex.ac.uk>
> 
> Prev by Date: st: Missing daily digest?
> Next by Date: Re: st: Missing daily digest?
> Previous by thread: RE: st: Beta values in QREG
> Next by thread: Re: st: metareg of failure rates or proportions / meta-
> analysis of proportions
> Index(es):
> Date
> Thread


*
*   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