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]

RES: st: pseudo panels implementation in Stata


From   "Henrique Neder" <hdneder@ufu.br>
To   <statalist@hsphsun2.harvard.edu>
Subject   RES: st: pseudo panels implementation in Stata
Date   Sun, 29 Apr 2012 11:03:19 -0300

Thanks for the answers. The complete citations of the papers are:
Verbeek, M. (2007). Pseudo panels and repeated cross-sections. Eletronic
copy available at:"http://ssrn.com/abstract=869445";.
Verbeek, M. and F.Vella (2005). Estimating Dynamic Models from Repeated
Cross-Sections, Journal of Econometrics, 127,83-102.   
Moffitt, R. (1993). Identification and Estimation of Dynamic Models with
Time Series of Repeated Cross-Sections, Journal of Econometrics, 59,99-123.
Mckenzie, D.J. (2004). Asymptotic Theory of Heterogeneous Dynamic
Pseudo-Panels, Journal of Econometrics, 120,235-262.

My repeated cross-section time series data is selected using a cluster
sample design. I agree that an important subject related to this problem is
the sample selection method of the repeated cross-section. Undoubtedly, this
will influence the standard errors estimates of the panel regressions: in
pseudo panels methods there are a measurement error problem. I know that for
this question perhaps there are no trivial answer, but ... How I can
integrate survey sample methods and panel estimation methods in Stata,
inside the pseudo panel context? I think that this is not a simple question
of data aggregation to perform cohort means. I believe there are numerous
problems related to obtaining consistent estimates as well as problems
concerning the adaptation of advanced techniques in dynamic panels (such as
obtaining robust standard errors, tests of over-identification restrictions,
auto-correlation errors tests, etc..) and applied to an estimation with
these averages cohorts. My question is: there are any developed empirical
research and procedures available referring to these problems? I would
greatly I would greatly appreciate any guidance on this appreciate any
guidance on this.    


Henrique Neder
Professor Associado - Instituto de Economia  
Universidade Federal de Uberlândia
(34) 32309564 
Av João Naves de Avila, 2121 Campus Santa Monica
38400-902  Uberlândia   MG 

     
   

-----Mensagem original-----
De: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Em nome de Richard Williams
Enviada em: Sunday, April 29, 2012 12:31 AM
Para: statalist@hsphsun2.harvard.edu; statalist@hsphsun2.harvard.edu
Assunto: Re: st: pseudo panels implementation in Stata

At 07:59 PM 4/28/2012, Henrique Neder wrote:
>Hello all,
>
>Has anybody knows any code in Stata that deals with pseudo panels? I am 
>trying to work with some dynamic panel models using repeated cross 
>section data that are cluster samples. I had read the important paper 
>from Verbeek ("Pseudo Panels and repeated cross-sections" and 
>"Estimating Dynamic Models from Repeated Cross-sections" ) but nothing 
>implemented empirically in this direction.  Surprisingly, it appears 
>that there are no much development in this field. An important question 
>discussed in these papers is  that pseudo panels are nothing more than 
>models with IV were the instruments are dummy variables referring to 
>the  cohorts.  In these sense my intuition is that the most of the 
>Stata resources  dealing with panel data models (and particularly with 
>dynamic panel models), like xtreg, xtabond, xtabond2,
>  could be used directed for pseudo panels, with the pertinent cautions.
>   Any comments are  welcome.

Others no doubt know much more about this than I do, but how does one create
a panel if each case is only measured at one point in time? 
Unless, say, you aggregated data up to the country level or something like
that.

My own impulse would be to include dummy variables for time and possibly
interactions for time, and then just use commands like regress and logit.
But, given that somebody has written papers on these topics - which, alas,
are kind of hard to look up because we don't have complete citations for
them (hint, hint) - maybe there are options I am not aware of.


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW:    http://www.nd.edu/~rwilliam

*
*   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/


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