[Openspace] question on weighted least square

[Julia Koschinsky]

Hi Allon,

It's currently not yet possible to control for both spatial autocorrelation
and heteroskedasticity in GeoDa. 

This is one of the topics in Dr. Anselin's summer workshop "Frontiers of
Spatial Regression"
(http://www.sal.uiuc.edu/news06/news-sum/frontiers-of-spatial-regression).
In the workshop labs, participants will be able to work with code that SAL
developed in this context. 

Best regards,
Julia

-----Original Message-----
From: openspace-bounces at sal.uiuc.edu [mailto:openspace-bounces at sal.uiuc.edu]
On Behalf Of Allon Crazy
Sent: Tuesday, March 06, 2007 11:39 AM
To: openspace at sal.uiuc.edu
Subject: [Openspace] question on weighted least square

I am a new user of Geoda. It seems my data needs to
put spatial weights. After I used the spatial lag and
error models, the heteroscedaticity is not gone,
still. I want to put the spatial weight back and use
GLS to run regression in my statistic software, but I
do not know how to view the spatial weights matched
with my orginal dataset so that I can open in my
regulat statistic software. Or, is there any way I can
calculate GLS in Geoda as well, in addition to
applying for spatial weights. 

Or any suggestions?

I would appreciate.


 
____________________________________________________________________________
________
Don't pick lemons.
See all the new 2007 cars at Yahoo! Autos.
http://autos.yahoo.com/new_cars.html 
_______________________________________________
Openspace mailing list
Openspace at sal.uiuc.edu
http://sal.uiuc.edu/mailman/listinfo/openspace