Quantcast

st: Sample selection in fixed effects model

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

st: Sample selection in fixed effects model

mw N. van Horen

Dear Statalisters

Does anyone know whether a user-written program is currently available in stata to estimate a sample selection model for panel data (using fixed effects). There was some discussion on the subject in http://www.stata.com/statalist/archive/2005-04/msg00112.html from which I understand it was not implemented in 2005. Does anyone know whether this has changed? Or is anyone familiar with a program written in another statistical package (matlab, gauss etc).

Thank you very much for your help.

Best,
Neeltje



De Nederlandsche Bank
Economic Policy and Research Division / Research Department
tel: +31-20-5245704

Access my research papers at:
http://www.dnb.nl/onderzoek/onderzoekers/persoonlijke-paginas/auto171982.jsp
http://ssrn.com/author=118741
http://ideas.repec.org/f/pva277.html


Please consider the environment before printing this email.

De informatie verzonden met dit e-mailbericht is vertrouwelijk en uitsluitend bestemd voor de geadresseerde. Indien u als niet-geadresseerde dit bericht ontvangt, wordt u verzocht direct de afzender hierover te informeren en het bericht te vernietigen. Gebruik van informatie door onbevoegden, openbaarmaking of vermenigvuldiging is verboden en kan leiden tot aansprakelijkheid. De afzender is niet aansprakelijk in geval van onjuiste overbrenging van het e-mailbericht en/of bij ontijdige ontvangst daarvan.

The information transmitted is confidential and intended only for the person or entity to whom or which it is addressed. If you are not the intended recipient of this communication, please inform us immediately and destroy this communication. Unauthorised use, disclosure or copying of information is strictly prohibited and may entail liability. The sender accepts no liability for improper transmission of this communication nor for any delay in its receipt.

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: st: Sample selection in fixed effects model

Maarten buis
--- On Mon, 1/2/10, mw N. van Horen wrote:
> Does anyone know whether a user-written program is
> currently available in stata to estimate a sample selection
> model for panel data (using fixed effects). There was some
> discussion on the subject in http://www.stata.com/statalist/archive/2005-04/msg00112.html
> from which I understand it was not implemented in 2005. Does
> anyone know whether this has changed? Or is anyone familiar
> with a program written in another statistical package
> (matlab, gauss etc).

From that post I read that the problem is that there is no such
thing as a fixed effects probit. That is still the case, as you
can read in the help file -xtprobit-:

    "There is no command for a conditional fixed-effects model,
     as there does not exist a sufficient statistic allowing
     the fixed effects to be conditioned out of thelikelihood.  
     Unconditional fixed-effects probit models may be fit with
     probit command with indicator variables for the panels.  
     However, unconditional fixed-effects estimates are biased."

From that I gather that there is not much hope that such a model
can ever be estimated regardless of software.

Sorry,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

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


     

*
*   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
|  
Report Content as Inappropriate

AW: st: Sample selection in fixed effects model

Martin Weiss-5

<>

" From that I gather that there is not much hope that such a model
can ever be estimated regardless of software."


Are you implying, then, that the two references in http://www.stata.com/statalist/archive/2005-04/msg00112.html cannot be implemented in Stata?



HTH
Martin


-----Urspr√ľngliche Nachricht-----
Von: [hidden email] [mailto:[hidden email]] Im Auftrag von Maarten buis
Gesendet: Montag, 1. Februar 2010 16:22
An: [hidden email]
Betreff: Re: st: Sample selection in fixed effects model

--- On Mon, 1/2/10, mw N. van Horen wrote:
> Does anyone know whether a user-written program is
> currently available in stata to estimate a sample selection
> model for panel data (using fixed effects). There was some
> discussion on the subject in http://www.stata.com/statalist/archive/2005-04/msg00112.html
> from which I understand it was not implemented in 2005. Does
> anyone know whether this has changed? Or is anyone familiar
> with a program written in another statistical package
> (matlab, gauss etc).

>From that post I read that the problem is that there is no such
thing as a fixed effects probit. That is still the case, as you
can read in the help file -xtprobit-:

    "There is no command for a conditional fixed-effects model,
     as there does not exist a sufficient statistic allowing
     the fixed effects to be conditioned out of thelikelihood.  
     Unconditional fixed-effects probit models may be fit with
     probit command with indicator variables for the panels.  
     However, unconditional fixed-effects estimates are biased."

>From that I gather that there is not much hope that such a model
can ever be estimated regardless of software.

Sorry,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

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


     

*
*   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
|  
Report Content as Inappropriate

Re: st: Sample selection in fixed effects model

mw N. van Horen
In reply to this post by Maarten buis

Dear Maarten

Thanks for your reply. There are ways around the problem that you mention. In the literature a number of solutions are mentioned, most often a Heckman two-stage method adapted to panel data or a ML method, which are applicable to fixed effects models (see also Hsiao (2003) Analysis of panel data page 236). However, I cannot find anyone using either one of these procedures in stata and I am hoping someone has done this.

Best,
Neeltje

De Nederlandsche Bank
Economic Policy and Research Division / Research Department
tel: +31-20-5245704

Access my research papers at:
http://www.dnb.nl/onderzoek/onderzoekers/persoonlijke-paginas/auto171982.jsp
http://ssrn.com/author=118741
http://ideas.repec.org/f/pva277.html


To

[hidden email]

cc


Subject

Re: st: Sample selection in fixed effects model

--- On Mon, 1/2/10, mw N. van Horen wrote:
> Does anyone know whether a user-written program is
> currently available in stata to estimate a sample selection
> model for panel data (using fixed effects). There was some
> discussion on the subject in
http://www.stata.com/statalist/archive/2005-04/msg00112.html
> from which I understand it was not implemented in 2005. Does
> anyone know whether this has changed? Or is anyone familiar
> with a program written in another statistical package
> (matlab, gauss etc).

From that post I read that the problem is that there is no such
thing as a fixed effects probit. That is still the case, as you
can read in the help file -xtprobit-:

   "There is no command for a conditional fixed-effects model,
    as there does not exist a sufficient statistic allowing
    the fixed effects to be conditioned out of thelikelihood.  
    Unconditional fixed-effects probit models may be fit with
    probit command with indicator variables for the panels.  
    However, unconditional fixed-effects estimates are biased."

From that I gather that there is not much hope that such a model
can ever be estimated regardless of software.

Sorry,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

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


     

*
*   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/



Please consider the environment before printing this email.

De informatie verzonden met dit e-mailbericht is vertrouwelijk en uitsluitend bestemd voor de geadresseerde. Indien u als niet-geadresseerde dit bericht ontvangt, wordt u verzocht direct de afzender hierover te informeren en het bericht te vernietigen. Gebruik van informatie door onbevoegden, openbaarmaking of vermenigvuldiging is verboden en kan leiden tot aansprakelijkheid. De afzender is niet aansprakelijk in geval van onjuiste overbrenging van het e-mailbericht en/of bij ontijdige ontvangst daarvan.

The information transmitted is confidential and intended only for the person or entity to whom or which it is addressed. If you are not the intended recipient of this communication, please inform us immediately and destroy this communication. Unauthorised use, disclosure or copying of information is strictly prohibited and may entail liability. The sender accepts no liability for improper transmission of this communication nor for any delay in its receipt.

pic06729.gif (1K) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: st: Sample selection in fixed effects model

Csizi89
This post has NOT been accepted by the mailing list yet.
Hi,

Were you able to solve this issue or find a program/code that does the job?
Loading...