s-news
[Top] [All Lists]

Testing data for the presence of Spatial Autocorrelation

To: <s-news@lists.biostat.wustl.edu>
Subject: Testing data for the presence of Spatial Autocorrelation
From: msc0238@geo.ed.ac.uk
Date: Wed, 26 Jun 2002 11:14:30 +0100
Organization: The University of Edinburgh
> Overview:
> I am trying to predict the distribution of black grouse in scotland using
GIS / S-Plus and
> the relevant environmental variable information. First I wish to test the
individual
> datasets for the presence of spatial autocorrelation to help me chose the
> type of modelling technique that would be appropriate.
> Initially I am testing grid data that I have imported into S-Plus spatial
from ASCII
> format and converted the dataframe to a matrix so as to maintain the two
> dimensional nature of the data.
>
> Questions:
> 1st
> The no. of returned Moran I values returned is equal to the number of
> columns that are within the grid. How can I get only one value returned
for the grid
> as a whole?
>
>
> 2nd
> Within the grids there are cells that contain no data (Null value), how
does
> s-Plus recognise these so as not to include the values within its
calculations?
>
> General Question:
Has anyone sucessfully used S-plus Spatial to test grid/polygon data for the
presence of spatial autocorrelation?
Any tips, help or suggestions gratefully accepted.
Many thanks
Judith



<Prev in Thread] Current Thread [Next in Thread>
  • Testing data for the presence of Spatial Autocorrelation, msc0238 <=