[R-sig-Geo] gstsls help

[Roger Bivand]

On Sun, 16 Jan 2011, Jennings Byrd wrote:

> All,
>
>
>
> I'm new to the R language and the packages therein.  What I wanted to know
> is if there are any way to obtain additional diagnostic info such as log
> likelihood or r square in the gstsls function in the spdep package.  I have
> searched the documentation and have largely come up empty handed.  Any help
> would be greatly appreciated.
>

Note that gstsls is a spatial TSLS, followed by a GLS, followed by a 
spatial TSLS. None of these are ML, so you don't have a likelihood 
expression, I think? I don't think that there is an easy expression for 
the R-squared either, other than the unadjusted ratio of the variance of 
the third step residuals and the variance of the response, but given the 
lag component, I'm not sure that this is appropriate either. This is not a 
well-explored area, contributions welcome.

Roger

>
>
> Will
>
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no