[Openspace] spatial models in STATA
koschins at uiuc.edu
Mon Nov 19 11:48:29 CST 2007
yes, they're available at: http://www.sal.uiuc.edu/stuff/stuff-sum/tutorials
---- Original message ----
>Date: Mon, 19 Nov 2007 18:40:44 +0100
>From: Jochen Mistelbacher <jocsar at t-online.de>
>Subject: Re: [Openspace] spatial models in STATA
>To: "German M. Izon" <g_mizon at yahoo.com>, openspace at sal.uiuc.edu
>Roger Bivand in his reply has already pointed to a previous post on the
>matter of getting weights from GeoDa to R. Additionally, there are some
>very very useful papers / workbooks available by Luc Anselin on weights
>and spatial regression in R, and of course Roger Bivand's description of
>the spdep-package. All of these together provide an excellent starting
>point for spatial analysis in R. Very much enjoyed it as a first-timer
>These are Luc Anselin's papers:
>Anselin, L. (2003): Data and Spatial Weights in spdep: Notes and
>Anselin, L. (2005): Spatial Regression Analysis in R: A Workbook.
>Probably available via the sal-homepage.
>Thank you Roger for your comments on the adjusted errors.
>German M. Izon wrote:
>> Thank you for your response. I am pretty new with R but I heard that I
>> can even run spatial 2-SLS in R. Is there a site that I can go to
>> learn about the R commands for spatial models. How can I import the
>> weight matrix in R?
>> Thank you!
>> */Jochen Mistelbacher <jocsar at t-online.de>/* wrote:
>> Hi German,
>> there are some .ado-files available by Maurizio Pisati, I think
>> called spatreg and spatwmat. I have not used them till now so can
>> say nothing much further. Tried once the weight-matrix-code but
>> had some memory problems in Stata. Alternatively one can probably
>> take a route via R, means importing the weight-matrix from GeoDa
>> into R and then re-exporting into Stata. But have not tried that
>> either. Anyway, if already that far one could just run all the
>> procedures in R as well.
>> In SHAC-errors I would be interested as well.
>> Greetings, Jochen
>> German Muchnik Izon wrote:
>>> Does anyone know if it is possible to run spatial models in STATA
>>> (i.e. lag, error, and 2SLS)? Can I export the weight matrix I
>>> create in GeoDA to STATA?
>>> I would also like to know if there is a way to calculate robust
>>> standard errors (SE) for spatial models? Is it similar to the
>>> White's robust SE for OLS models?
>>> Thank you!
>>> University of New Mexico.
>>> Openspace mailing list
>>> Openspace at sal.uiuc.edu
>> Be a better sports nut! Let your teams follow you with Yahoo Mobile.
>> Try it now.
>Openspace mailing list
>Openspace at sal.uiuc.edu
More information about the Openspace