The purpose of adopting this approach using maps and error processes with simple but well defined properties is to understand better how different elements of the situation, individually and together, contribute to the final propagated error.
How might the propagated error impact on the purpose of any analysis such as the delimitation of deprived areas or the identification of hot spots or outliers?
The presence of location error interacting with the spatial structure in the source maps, the presence of spatial correlation in the errors of the attribute measurement process, or indeed their simultaneous presence, are all capable of generating spatially complex maps of propagated error under these operations.
Each is associated with different spatial correlation gradients in source maps, so their contribution to the propagated error structure will vary with the strength of source-map correlation at different lags, as noted earlier for Figures 2a and 3a.
e](1,0) denotes the (1,0) lag spatial correlation in the propagated error and MSE is the mean square error.
I find the log exceedingly helpful there, as when I see many propagated errors
, and an incorrect level of variables and/or observations, I immediately know that extra data or unincluded factor levels were popping up.