For the spatial aggregations by district, we should NOT weight values by area. It should be the mean pixel value, i.e., the mean anomaly per pixel area for the district. Each pixel is weighted the same.
We'll include the polygon areas as weights in the model as offsets, to represents the probability of outbreak per size of the district.
For the spatial aggregations by district, we should NOT weight values by area. It should be the mean pixel value, i.e., the mean anomaly per pixel area for the district. Each pixel is weighted the same.
We'll include the polygon areas as weights in the model as offsets, to represents the probability of outbreak per size of the district.