estimation of spatial panels with endogenous weighting matrices

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

estimation of spatial panels with endogenous weighting matrices

merveaksoylar
Dear all,

I'm doing my thesis on some empirical work on spatial panel data.
I already use some geographical weighting matrices (using spdep and splm
packages).

But now I'm planning to use some endogenous weighting matrices, which are
time variant.

Could you please help me for this estimation procedure? Do you suggest any
packages for endogenous W, or is it possible to send me some codes for this
kind of estimation?

Thanks a lot,
Merve Aksoylar

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Reply | Threaded
Open this post in threaded view
|

Re: estimation of spatial panels with endogenous weighting matrices

Roger Bivand
Administrator
On Thu, 2 Nov 2017, MERVE AKSOYLAR wrote:

> Dear all,
>
> I'm doing my thesis on some empirical work on spatial panel data.
> I already use some geographical weighting matrices (using spdep and splm
> packages).
>
> But now I'm planning to use some endogenous weighting matrices, which are
> time variant.

Please provide a working example if possible (time invariant, error or
lag, fixed or random effects?), or at least references to published
methods articles. Why might time variant endogenous weighting matrices
(size T x NxN x T with many 0 by definition) be interesting? You are
thinking of a different Kronecker product than that in splm, but with what
motivation (theoretical and empirical)? How are you thinking of proceeding
- GMM, ML, or Bayesian? Could you use INLA in a non-separable setting?
This doesn't get you to endogenous weights, though.

Roger

>
> Could you please help me for this estimation procedure? Do you suggest any
> packages for endogenous W, or is it possible to send me some codes for this
> kind of estimation?
>
> Thanks a lot,
> Merve Aksoylar
>
> [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> [hidden email]
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>

--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: [hidden email]
Editor-in-Chief of The R Journal, https://journal.r-project.org/index.html
http://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Roger Bivand
Department of Economics
Norwegian School of Economics
Helleveien 30
N-5045 Bergen, Norway
Reply | Threaded
Open this post in threaded view
|

Re: estimation of spatial panels with endogenous weighting matrices

merveaksoylar
Actually my research is on economic growth theory, and I'm doing some
empirical work on Emerging Market Economies (24 countries) between
1991-2016 period.

For searching the growth spillovers, I've used the geographical distance
weighting matrix (exogenous) to estimate my growth regression. (SAR with FE)
But I decided to change the definition of space for Emerging Markets,
because they don't have exact similarities on geographical basis. Therefore
I'm planning to create an institutional weighting matrix for my estimation
and it is time variant, which makes also an endogeneity problem too.
Do you suggest me any estimation method? I was planning to do it again on
ML estimation, but maybe Bayesian will work better? But I don't know how to
do it. Could you please help me.

Thanks,
Merve Aksoylar

On Thu, Nov 2, 2017 at 3:13 PM, Roger Bivand <[hidden email]> wrote:

> On Thu, 2 Nov 2017, MERVE AKSOYLAR wrote:
>
> Dear all,
>>
>> I'm doing my thesis on some empirical work on spatial panel data.
>> I already use some geographical weighting matrices (using spdep and splm
>> packages).
>>
>> But now I'm planning to use some endogenous weighting matrices, which are
>> time variant.
>>
>
> Please provide a working example if possible (time invariant, error or
> lag, fixed or random effects?), or at least references to published methods
> articles. Why might time variant endogenous weighting matrices (size T x
> NxN x T with many 0 by definition) be interesting? You are thinking of a
> different Kronecker product than that in splm, but with what motivation
> (theoretical and empirical)? How are you thinking of proceeding - GMM, ML,
> or Bayesian? Could you use INLA in a non-separable setting? This doesn't
> get you to endogenous weights, though.
>
> Roger
>
>
>> Could you please help me for this estimation procedure? Do you suggest any
>> packages for endogenous W, or is it possible to send me some codes for
>> this
>> kind of estimation?
>>
>> Thanks a lot,
>> Merve Aksoylar
>>
>>         [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> R-sig-Geo mailing list
>> [hidden email]
>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>>
>>
> --
> Roger Bivand
> Department of Economics, Norwegian Sc
> <https://maps.google.com/?q=Norwegian+Sc&entry=gmail&source=g>hool of
> Economics,
> Helleveien 30, N-5045 Bergen, Norway.
> voice: +47 55 95 93 55; e-mail: [hidden email]
> Editor-in-Chief of The R Journal, https://journal.r-project.org/index.html
> http://orcid.org/0000-0003-2392-6140
> https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
>

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo