[R-sig-Geo] How to use R scripts to estimate two-weight matrix SAR models

[Roger Bivand]

On Sun, 18 Oct 2015, Chen Lee wrote:

> Seek expert help: How to use R scripts to estimate two-weight matrix SAR
> models? The two-weight matrix SAR model is specified as: Y=ρ1W1Y+ρ2W2Y+Xβ+ε.
>
> 	[[alternative HTML version deleted]]

Please do not post HTML.

There are no functions in R for fitting models of this kind. Further, it 
would be important to recall that the fitted coefficients on the X 
variables should not be interpreted directly - rather through their 
impacts, and methods using traces do not (yet) exist for computing such 
impacts. So even if you try to implement a function for model fitting, you 
would also need to implement a suitable impacts method. I am not aware of 
any available implementations in any language, although there may be 
something somewhere. So first the methods would need to be elaborated, the 
inplemented and tested.

Roger

>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at 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; fax +47 55 95 91 00
e-mail: Roger.Bivand at nhh.no