I want to develop a multivariate logistic regression to model the odds of opioid overdose deaths at the ZIP Code level.
Since the logistic regression does not handle spatial autocorrelation, I need to check whether my dependent/outcome variable is spatially autocorrelated or not (measuring the Moran’s I on the occurrences of opioid overdose death).
I wonder if the outcome was spatially autocorrelated, how I can address that in ArcGIS Pro?
I know that people use the Bayesian hierarchical spatial model to account for spatial autocorrelation but how can I use such a method in ArcGIS?
Does the GWR method account for spatial autocorrelation?
