[R-sig-Geo] unbalanced panel

[Roger Bivand]

On Thu, 30 Aug 2018, Jérémie Juste wrote:

> Hello,
>
> I was wondering if there are any plans to allow the estimation of
> unbalanced panel in spdep or splm. I am not fortunate enough to have
> balanced spatial data and I think this woud be a valuable input to the
> community.

Could you please give us more motivation - are the missing observations 
patterned? Which application area are you considering? What literature are 
you using with regard to estimating unbalanced panels - is the problem 
more temporal than spatial? Is this actually about interpolating or 
imputing the missing values and the carrying through uncertainty? If this 
was the case, then a Bayesian approach might be desirable, to permit the 
imputation uncertainty to be carried through?

>
> Is there a way to compute direct and indirect effects when estimating an
> unbalanced panel spatial durbin model ?

First fit the model! Probably, unless you have a solid micro-model for why 
global spillover is needed, you should keep the spatial process in the 
error (local spillover), simplifying the impacts. In any case, avoiding 
econometric fixed effects may let you use statistical random effects 
models with (separable) temporal and spatial structure.

Roger

>
> Best regards,
> Jeremie Juste
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo using r-project.org
> 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: Roger.Bivand using nhh.no
http://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en