Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: 3SLS with forward differencing
From
Anirudh Tagat <[email protected]>
To
statalist <[email protected]>
Subject
Re: st: 3SLS with forward differencing
Date
Wed, 27 Nov 2013 16:20:50 +0530
> -----Original Message-----
> From: [email protected] [mailto:owner-
> [email protected]] On Behalf Of Anirudh Tagat
> Sent: 25 November 2013 12:34
> To: statalist
> Subject: Re: st: 3SLS with forward differencing
>
> Dear Billy,
>
> My apologies for not following the statalist rules more precisely. I am hopeful
> that this reply conforms more closely to the list
> rules/etiquette:
>
> I am currently trying to run an estimation following Mattias Lundberg and Lyn
> Squire, "The Simultaneous Evolution of Growth and Inequality", The Economic
> Journal , Vol. 113, No. 487 (Apr., 2003), pp. 326-344) -- where the efficient
> estimation methodology used was derived from Keane and Runkle, "On the
> Estimation of Panel-Data Models with Serial Correlation When Instruments Are
> Not Strictly Exogenous", Journal of Business & Economic Statistics , Vol. 10, No.
> 1 (Jan., 1992), pp. 1-9.
>
> I am using Stata 12.1, and dealing with dynamic panel data similar to that
> commonly found in cross-country analyses (4 periods, across 200 observations),
> and would like to use first-differencing (alternatively forward differencing) in
> line with the above mentioned studies. So far, I have not been clear on how this
> is best achieved, and if the
> -reg3- or -ivreg2- (or -xtabond2- (David Roodman, 2003. "XTABOND2:
> Stata module to extend xtabond dynamic panel data estimator,"
> Statistical Software Components S435901, Boston College Department of
> Economics, revised 22 Sep 2013.), which is close but does not follow Keane and
> Runkle) function is indeed the appropriate tool. Any suggestions on how to run
> this estimation (or at least implementing the transformation) using Stata would
> be most welcome.
Keane-Runkle is a kind of panel GLS estimator that assumes a form
serial correlation that is invariant across panels. A short and clear
description of KR and how to implement it is in Baltagi's "Econometric
Analysis of Panel Data" (I have the 4th edition at hand, and it's on
p. 162). Another simple and clear description of how to do it is in
Baltagi's chapter in Nigar Hashimzade, Michael A. Thornton (eds),
Handbook of Research Methods and Applications in Empirical
Macroeconomics, pp. 240-1.
As far as I know, no one has implemented this in Stata, so you'd need
to do it by hand yourself. But based on Baltagi's explanation, it
doesn't look too bad.
HTH,
Mark
Many thanks for your suggestion, Mark. When you say 'by hand' I
suppose you are referring to making changes to the data manually,
which is what I figured would be the only way, but since I am unsure
of where to start with such a transformation, I was hoping to find
some pointers on where exactly (or how exactly) to perform these
operations in Stata. With that said, I am looking at Baltagi's notes
on methodology, and have RATS (Estima) code available with an example
but since I am not familiar with RATS, the code is proving to be
increasingly difficult to interpret in Stata terms. Typing -xtabond2-:
xtabond2 delta_growth delta_gini gini71 ln_inc71 res weather
banking_index devindex ln_vexp percentSC82 edu71 ker
ala-tamilnadu, iv(landgini71#land_reform#land dist_bus71 dist_market
dist_dhq predicted_conflict overlap) robust twostep
*! Favoring space over speed. To switch, type or click on mata: mata
set matafavor speed, perm.
*! No observations.
*! r(2000);
>
> Many thanks
> Anirudh Tagat
> *
> * 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/
- -----
Sunday Times Scottish University of the Year 2011-2013
Top in the UK for student experience
Fourth university in the UK and top in Scotland (National Student Survey 2012)
We invite research leaders and ambitious early career researchers to
join us in leading and driving research in key inter-disciplinary themes.
Please see www.hw.ac.uk/researchleaders for further information and how
to apply.
Heriot-Watt University is a Scottish charity
registered under charity number SC000278.
*
* 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/