st: analytic standard errors in quantile regression

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

st: analytic standard errors in quantile regression

Jochen_Spaeth@t-online.de
Dear all,

I have a large panel data set (N >> T ) and I want to run quantile
regressions with heteroskedasticity/autocorrelation-robust standard
errors. However, there is no such option that could be passed to the
-qreg- command in STATA and because my data set is huge bootstrap
standard errors do not seem a viable alternative. Is anyone out there
aware of a program that defines such a -robust- option for the -qreg
command- or something similar?

Any comment is highly appreciated!

Sincerely,
Jochen


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
Reply | Threaded
Open this post in threaded view
|

st: RE: analytic standard errors in quantile regression

Martin Weiss-5

Line for the server...

***********
-findit clad-
***********

HTH
Martin


-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of
[hidden email]
Sent: Wednesday, December 03, 2008 11:16 AM
To: [hidden email]
Subject: st: analytic standard errors in quantile regression

Dear all,

I have a large panel data set (N >> T ) and I want to run quantile
regressions with heteroskedasticity/autocorrelation-robust standard
errors. However, there is no such option that could be passed to the
-qreg- command in STATA and because my data set is huge bootstrap
standard errors do not seem a viable alternative. Is anyone out there
aware of a program that defines such a -robust- option for the -qreg
command- or something similar?

Any comment is highly appreciated!

Sincerely,
Jochen


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/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/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
Reply | Threaded
Open this post in threaded view
|

Re: st: RE: analytic standard errors in quantile regression

Jochen_Spaeth@t-online.de
Hello Martin,

thank you very much for your comment. However, STATA's -clad- command is
not precisely what I am looking for since it relies on bootstrap
standard errors, too, which I could have performed with -bsqreg-, too.
Rather than that, I am looking for analytic standard errors for the
heteroskedasticity case.

Jochen

-----Original Message-----
Date: Wed, 03 Dec 2008 11:20:26 +0100
Subject: st: RE: analytic standard errors in quantile regression
From: "Martin Weiss" <[hidden email]>
To: <[hidden email]>


Line for the server...

***********
-findit clad-
***********

HTH
Martin


-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of
[hidden email]
Sent: Wednesday, December 03, 2008 11:16 AM
To: [hidden email]
Subject: st: analytic standard errors in quantile regression

Dear all,

I have a large panel data set (N >> T ) and I want to run quantile
regressions with heteroskedasticity/autocorrelation-robust standard
errors. However, there is no such option that could be passed to the
-qreg- command in STATA and because my data set is huge bootstrap
standard errors do not seem a viable alternative. Is anyone out there
aware of a program that defines such a -robust- option for the -qreg
command- or something similar?

Any comment is highly appreciated!

Sincerely,
Jochen


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/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/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/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
Reply | Threaded
Open this post in threaded view
|

st: R: analytic standard errors in quantile regression

Carlo Lazzaro
In reply to this post by Jochen_Spaeth@t-online.de
Dear Jochen,
some threads ago, the following textbook was quoted on the list:

Roger Koenker. Quantile Regression. Cambridge University Press, 2005
(Paperback).

Unfortunately, since I have ordered some days ago, I do not know whether or
not its contents tackle your problem.

Kind Regards,
Carlo

-----Messaggio originale-----
Da: [hidden email]
[mailto:[hidden email]] Per conto di
[hidden email]
Inviato: mercoledì 3 dicembre 2008 11.16
A: [hidden email]
Oggetto: st: analytic standard errors in quantile regression

Dear all,

I have a large panel data set (N >> T ) and I want to run quantile
regressions with heteroskedasticity/autocorrelation-robust standard
errors. However, there is no such option that could be passed to the
-qreg- command in STATA and because my data set is huge bootstrap
standard errors do not seem a viable alternative. Is anyone out there
aware of a program that defines such a -robust- option for the -qreg
command- or something similar?

Any comment is highly appreciated!

Sincerely,
Jochen


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/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/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
Reply | Threaded
Open this post in threaded view
|

st: RE: R: analytic standard errors in quantile regression

Martin Weiss-5
Line for the server...

There is only one entry in the subject index for heteroscedasticity, and a
cursory look at it does not bode well for Jochen`s problem...

HTH
Martin

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Carlo Lazzaro
Sent: Wednesday, December 03, 2008 2:03 PM
To: [hidden email]
Cc: [hidden email]
Subject: st: R: analytic standard errors in quantile regression

Dear Jochen,
some threads ago, the following textbook was quoted on the list:

Roger Koenker. Quantile Regression. Cambridge University Press, 2005
(Paperback).

Unfortunately, since I have ordered some days ago, I do not know whether or
not its contents tackle your problem.

Kind Regards,
Carlo

-----Messaggio originale-----
Da: [hidden email]
[mailto:[hidden email]] Per conto di
[hidden email]
Inviato: mercoledì 3 dicembre 2008 11.16
A: [hidden email]
Oggetto: st: analytic standard errors in quantile regression

Dear all,

I have a large panel data set (N >> T ) and I want to run quantile
regressions with heteroskedasticity/autocorrelation-robust standard
errors. However, there is no such option that could be passed to the
-qreg- command in STATA and because my data set is huge bootstrap
standard errors do not seem a viable alternative. Is anyone out there
aware of a program that defines such a -robust- option for the -qreg
command- or something similar?

Any comment is highly appreciated!

Sincerely,
Jochen


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/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/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/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
Reply | Threaded
Open this post in threaded view
|

Re: st: R: analytic standard errors in quantile regression

Stas Kolenikov
In reply to this post by Carlo Lazzaro
It does (section 3.4). I don't really know how you would take into
account the panel nature of your data though. At the very least, your
sandwich estimator (that's what's discussed in Koenker's book in that
section) should be summing over panels, not over individual
observations. I would seriously doubt anybody has implemented it.

On 12/3/08, Carlo Lazzaro <[hidden email]> wrote:

> Dear Jochen,
>  some threads ago, the following textbook was quoted on the list:
>
>  Roger Koenker. Quantile Regression. Cambridge University Press, 2005
>  (Paperback).
>
>  Unfortunately, since I have ordered some days ago, I do not know whether or
>  not its contents tackle your problem.
>
>  Kind Regards,
>  Carlo
>
>  -----Messaggio originale-----
>  Da: [hidden email]
>  [mailto:[hidden email]] Per conto di
>  [hidden email]
>  Inviato: mercoledì 3 dicembre 2008 11.16
>  A: [hidden email]
>  Oggetto: st: analytic standard errors in quantile regression
>
>  Dear all,
>
>  I have a large panel data set (N >> T ) and I want to run quantile
>  regressions with heteroskedasticity/autocorrelation-robust standard
>  errors. However, there is no such option that could be passed to the
>  -qreg- command in STATA and because my data set is huge bootstrap
>  standard errors do not seem a viable alternative. Is anyone out there
>  aware of a program that defines such a -robust- option for the -qreg
>  command- or something similar?
>
>  Any comment is highly appreciated!
>
>  Sincerely,
>  Jochen
>
>
>  *
>  *   For searches and help try:
>  *   http://www.stata.com/help.cgi?search
>  *   http://www.stata.com/support/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/statalist/faq
>  *   http://www.ats.ucla.edu/stat/stata/
>


--
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
Reply | Threaded
Open this post in threaded view
|

Re: st: RE: R: analytic standard errors in quantile regression

Jochen_Spaeth@t-online.de
In reply to this post by Martin Weiss-5
Dear Carlo,

I do not know Koenker's (2005) book. However, Hao and Naiman (2007,
p.44ff.) derive analytic standard errors both for the iid and the
non-idd case. Thus, at least in theory the problem is already solved.

Kind Regards,
Jochen




-----Original Message-----
Date: Wed, 03 Dec 2008 14:08:47 +0100
Subject: st: RE: R: analytic standard errors in quantile regression
From: "Martin Weiss" <[hidden email]>
To: <[hidden email]>

Line for the server...

There is only one entry in the subject index for heteroscedasticity, and
a
cursory look at it does not bode well for Jochen`s problem...

HTH
Martin

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Carlo Lazzaro
Sent: Wednesday, December 03, 2008 2:03 PM
To: [hidden email]
Cc: [hidden email]
Subject: st: R: analytic standard errors in quantile regression

Dear Jochen,
some threads ago, the following textbook was quoted on the list:

Roger Koenker. Quantile Regression. Cambridge University Press, 2005
(Paperback).

Unfortunately, since I have ordered some days ago, I do not know whether
or
not its contents tackle your problem.

Kind Regards,
Carlo

-----Messaggio originale-----
Da: [hidden email]
[mailto:[hidden email]] Per conto di
[hidden email]
Inviato: mercoledì 3 dicembre 2008 11.16
A: [hidden email]
Oggetto: st: analytic standard errors in quantile regression

Dear all,

I have a large panel data set (N >> T ) and I want to run quantile
regressions with heteroskedasticity/autocorrelation-robust standard
errors. However, there is no such option that could be passed to the
-qreg- command in STATA and because my data set is huge bootstrap
standard errors do not seem a viable alternative. Is anyone out there
aware of a program that defines such a -robust- option for the -qreg
command- or something similar?

Any comment is highly appreciated!

Sincerely,
Jochen


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/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/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/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/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/