Hi,
thanks for the link. Actually, I have already that presentation. The last line on page 30 refers to GMM estimation, while spatreg command relies on ML estimation. In particular, I read a paper where ML was used to estimate a spatial lag model on panel data which is not that different from my approach. That was why I was wondering whether there was a way to use that command on panel data?
Best regards
Phil
-------- Original-Nachricht --------
> Datum: Wed, 18 Feb 2009 11:45:35 -0500
> Von: Austin Nichols <[email protected]>
> An: [email protected]
> Betreff: Re: st: Spatial estimation techniques on panel data - spatreg possible?
> See
> http://www.stata.com/meeting/snasug08/drukker_spatial.pdf
> esp. the last line on page 30.
>
> On Wed, Feb 18, 2009 at 11:10 AM, <[email protected]> wrote:
> > Hello,
> >
> > I am new on the subscriber list. I currently try to estimate a model
> explaining bilateral multinational activity which is augmented by a spatial
> lag. My problem is that I want to use panel data and that the spatreg
> commands only work for cross-sections. I saw an earlier question in this context
> on Statlist which was not further treated due to the lack of theory.
> However, there are a number of economic papers which have already applied spatial
> estimation techniques on panel data.
> >
> > The basic problem is that the weighting matrix is constructed according
> to the observations of one year only. I am now wondering whether it is
> possible to somehow tell STATA that the vector of the dependent variable should
> be multiplied with the weighting matrix for each year seperately?
> >
> > Another possibility would be to manipulate the data in this way before
> loading it to STATA. However, then it is obviously not possible to use the
> spatreg commands anymore.
>
> *
> * 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/
--
Jetzt 1 Monat kostenlos! GMX FreeDSL - Telefonanschluss + DSL
für nur 17,95 Euro/mtl.!* http://dsl.gmx.de/?ac=OM.AD.PD003K11308T4569a
*
* 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/
