Hi Luc,
I have been using GEODA to compute bivariate Moran's I to examine
spatial correlations between various plant and soil variables I have
collected. I am favouring the use of a bivariate Moran's I over
cross-correlograms as my plots are very small (5m x 5m) and there is not
much spatial structure on this scale in my system. It is, therefore,
nice to be able to use the threshold weighting capability in GEODA to
compare values of one variable with another over certain threshold
distances, rather than using spatial lag distances. Could you perhaps
point me in the direction of a reference that gives explicit methods for
how GEODA makes these computations? (i.e. does GEODA use Wartenberg's
bivariate Moran's I? how are points near the sample point weighted
(1/d?) etc).
Thanks so much
Kate
Kate Searle
Post Doctoral Fellow
CSIRO - Sustainable Ecosystems
Davies Lab
PMB PO Aitkenvale
Queensland 4814
Australia
kate.searle@csiro.au
61 7 4753 8564
We shall not cease from exploration
And the end of all our exploring
Will be to arrive where we started
And know the place for the first time
T. S. Elliot
[[alternative HTML version deleted]]