Serendipitously, I was just looking for the same or very similar sort of
discussion just yesterday...
First a bit of truly shameless (and shameful!) self-promotion: Jude Hays
& I had a little (HIGHLY preliminary) monte carlo on mis-specification
of W (you say M, we've always seen W used--tomayto, tomahto--but you
mean the matrix of connectivity weights, yes?)... It's on pages 100-105
here (copied from some of our very earliest papers on spatial models):
http://www-personal.umich.edu/~franzese/FranzeseHays.DraftOutlineToC.Spa
tialEconometricModelsInterdependence.pdf
With the obligatory self-promotion now out of the way, what you really
seem to want is a good discussion of the issues &, maybe, some recent
thoughts on how to help avoid some of the problems that may attend such
mis-specification. Here's one (of many, no doubt) recent better place to
start:
Constructing the Spatial Weights Matrix Using a Local Statistic.
A Getis, J Aldstadt - Geographical Analysis, 2004
http://muse.jhu.edu/journals/geographical_analysis/v036/36.2getis.pdf
There were also at least two papers on the matter, or related, in last
year's conference of the spatial econometric association:
Specification of Spatial Models: Analysis, Comparisons, Suggestions
Stanislav Stakhovych & Tammo H.A. Bijmolt
http://fp.paceprojects.f9.co.uk/Stakhovych.pdf
Taking off some hoods: estimating spatial models with a non-arbitary W
matrix
Esteban Fernandez-Vazquez & Jorge Rodriguez-Valez
http://fp.paceprojects.f9.co.uk/Vazquez.pdf
(The title is a nod to one of Anselin's well-cited reviews.)
Hope this helps,
Rob
************************************************************************
*
Robert (Rob) J. Franzese, Jr. US Mail: (Room 4246
ISR)
Associate Professor, P.O. Box
1248
Department of Political Science, Ann Arbor, MI
48106-1248
and Research Associate Professor, (Currier: 426 Thompson
St.)
Center for Political Studies, e-mail:
[log in to unmask]
Institute for Social Research, office:
1-734-936-1850
The University of Michigan fax:
1-734-764-3341
http://www-personal.umich.edu/~franzese
************************************************************************
*
-----Original Message-----
From: Political Methodology Society [mailto:[log in to unmask]] On
Behalf Of [log in to unmask]
Sent: Thursday, July 24, 2008 10:33 AM
To: [log in to unmask]
Subject: [POLMETH]
I am currently looking for a good explination for the problem of
miss-specifying a M matrix in a Spatial Regression. Does anyone
know of a good article that adresses this problem?
TL Long
Full Time Instructor
Department of Liberal Arts
Temple College
[log in to unmask]
**********************************************************
Political Methodology E-Mail List
Editors: Melanie Goodrich, <[log in to unmask]>
Delia Bailey, <[log in to unmask]>
**********************************************************
Send messages to [log in to unmask]
To join the list, cancel your subscription, or modify
your subscription settings visit:
http://polmeth.wustl.edu/polmeth.php
**********************************************************
**********************************************************
Political Methodology E-Mail List
Editors: Melanie Goodrich, <[log in to unmask]>
Delia Bailey, <[log in to unmask]>
**********************************************************
Send messages to [log in to unmask]
To join the list, cancel your subscription, or modify
your subscription settings visit:
http://polmeth.wustl.edu/polmeth.php
**********************************************************
|