Answers are grounded in this paper and cite their sources.
In modeling the AIDS epidemic, the geographic dimensions have been totally ignored, partly as a result of ignorance, partly from an exaggerated and unreasoned concern for confidentiality. For both educational intervention (animated cartography for television) and health planning (expanding already overstressed facilities), it is necessary to predict the next maps, not just numbers down the time line. Spatial adaptive filtering, combined with parametric tracking, searches out and employs both spatial and temporal structure in (x, y, t) series, producing geographic forecasts with greatly reduced error terms compared to conventional approaches.
No takes yet. Share an insight, caveat, or question.
Gould et al. (1991) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: