The expected value of Moran's I is given in the literature as (-1/[N-1]), where N=Number of zones. In R software, when Moran's I for a response variable is calculated, it is same as (-1/[N-1]), but it is something else for Moran's I of regression model residuals with the same data. Why is this happening? Does anyone have any clue?

More Sowjanya Dhulipala's questions See All
Similar questions and discussions