[R-sig-Geo] Lagrange multiplier tests for weighted least squares

Roger Bivand Roger.Bivand at nhh.no
Wed Apr 21 15:19:25 CEST 2010


On Wed, 21 Apr 2010, Roberto Patuelli wrote:

> Hi everyone,
>
> The help file for the spdep command "lm.LMtests" for running Lagrange 
> multiplier tests per spatial dependence suggests not to use weights in the lm 
> object given as argument.
> If weights are highly relevant for a certain model (for example, because of 
> strong heterogeneity in the size/importance of the regions being analysed), 
> is there a way to get around this limitation? I struggle to find anything 
> online relating weighted least squares and LM tests for spatial dependence.

I think econometricians, spatial or not, typically do not "do" weights, 
but prefer to use HAC or similar standard errors. LM tests are fro the 
spatial econometrics approach, so they do not accommodate weights. To add 
them would mean doing the math to see where in the development the omega 
should be added into the spatial process, then one would need to show that 
the tests perform as expected. Nice paper if one had time!

Roger

>
> Thanks for the help.
> Roberto
>
> ********************
> Roberto Patuelli, Ph.D.
> Istituto Ricerche Economiche (IRE) (Institute for Economic Research)
> Università della Svizzera Italiana (University of Lugano)
> via Maderno 24, CP 4361
> CH-6904 Lugano
> Switzerland
> Phone: +41-(0)58-666-4166
> Fax: +39-02-700419665
> Email: roberto.patuelli at usi.ch
> Homepage: http://www.people.lu.unisi.ch/patuellr
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> 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


More information about the R-sig-Geo mailing list