Moran's I explanation: "It's important to understand what "nonstationary" means, Todd. It is a property of a stochastic model of a phenomenon, not the phenomenon itself. At each location within the extent of a spatial field, a stochastic model specifies a probability distribution for the field's value there. It also specifies a bivariate distribution of the values at any pair of points, a trivariate distribution of the values at any three points, and so on. The model is stationary when those distributions do not change when you uniformly translate all locations.
"
'via Blog this'
No comments:
Post a Comment