[Openspace] Multiple observations per location in GeoDa

[Josh Woodard]

Sorry, I just realized I sent the wrong link.  Here is the correct one.

Use the following links to access the corresponding files:

/jdwoodar/Spatial/MultiWeightsExample.zip:
https://netfiles.uiuc.edu/xythoswfs/webui/_xy-25926472_2-t_VLuvBPl5


----- Original Message ----- 
From: "Josh Woodard" <jdwoodar at uiuc.edu>
To: "Graeme Byrne" <G.Byrne at latrobe.edu.au>; <openspace at sal.uiuc.edu>
Sent: Friday, November 16, 2007 8:56 PM
Subject: Re: [Openspace] Multiple observations per location in GeoDa


> Graeme,
>
>
>
> You may find the following files useful.  I had experimented with 
> something similar in the past so I had this laying around.  In the file 
> "CreateMultiLocWeights" you will find VBA code to do what you describe 
> (open the VBA editor to view the code).  It's not pretty, but should 
> provide a one time solution.  Basically, what it does is creates GAL 
> weights for individual observations when only the area the observations 
> are in are known.  Each observation in each area will be a neighbor of 
> each observation in that area, and also all observations in neighboring 
> areas will become neighbors as well.  You need the shapefile for the areas 
> or the GAL file for the area weights.  Obviously you won't get a shapefile 
> for the individual observations, but you can do reg's with just the dbf 
> and weights in GeoDa. You can create it using this code in Excel '07 as 
> long as you don't have more than about 600,000 or so observations (or 
> about 32,000 if you don't have Excel '07).
>
>
>
> I am not sure how "appropriate" this is statistically, so I would not be 
> the one to consult on that, but this should help you in creating such a 
> weights file which can be used in GeoDa.  I would be interested to hear 
> anyone else's thoughts on such a design though as I ran across a similar 
> problem.  I was able to implement it fine, but wasn't quite sure if this 
> was a viable approach.
>
>
>
> There are some short instructions and an example to show you how to 
> convert the files back and forth.  Please let me know if you have any 
> questions or run into any problems. Let me know if you have GWT weights or 
> need to GWT weights...I have some code for that too.
>
>
>
> Josh Woodard
>
>
>
>
> Use the following links to access the corresponding files:
>
> /jdwoodar/Spatial/MultiWeightsExample.zip:
> https://netfiles.uiuc.edu/xythoswfs/webui/_xy-25926587_2-t_Er3YgGgx
>
>
>
> -----------------------------------------------------------------------------
> Joshua D. Woodard
> Ph.D. Candidate
> Office for Futures and Options Research (OFOR)
> Farmdoc-Finance, Insurance, and Risk Management Group
> Department of Agricultural and Consumer Economics
> University of Illinois at Urbana-Champaign
>
>
>
>
> ----- Original Message ----- 
> From: "Graeme Byrne" <G.Byrne at latrobe.edu.au>
> To: <openspace at sal.uiuc.edu>
> Sent: Saturday, November 10, 2007 9:03 PM
> Subject: [Openspace] Multiple observations per location in GeoDa
>
>
> Hi all, I have spatial regression problems where there are multiple 
> observations per locations. One particular model has cases for males and 
> females from the same place and I want to estimate the effect of gender 
> (via a dummy variable) as well as the effect of some gender specific 
> covariates. I can create a shape file from MapInfo but GeoDa requires a 
> unique location for each case so at the moment I am fitting separate 
> models for males and females which works but reduces the power of 
> statistical tests. Ideally I would like to include gender and possibly 
> other factors (e.g. Age) in a single SAR or SE model. Are there R routines 
> available or can GeoDa be coerced into doing this
>
> Graeme
>
>