Moran's I explanation

Discussion created by toddmcd on Dec 29, 2010
Latest reply on Feb 20, 2013 by peter_m
I am using the Moran's I tool to check for spatial autocorrelation in the residuals generated from the OLS model to determine if geographically weighted regression might be appropriate for the data.

I am wondering what the Moran's I tool uses to calculate the z-score? I have read the documentation and it seems that it determines the difference between each residual value and the mean of all the residuals, and then checks to see if these magnitudes are clustered or dispersed (or random).

Can someone verify my interpretation and/or provide additional explanation?