Quantifying Aggregated Uncertainty in Plasmodium falciparum Malaria Prevalence and Populations at Risk via Efficient Space-Time Geostatistical Joint Simulation
Figure 3
Simulated global surfaces of PfPR2–10.
Examples of five of the 500 realisations of PfPR2–10 generated via the joint simulation algorithm. Each of these maps represents an equally possible ‘reality’ and the full set of 500 provides a model of the probable prevalence at all locations. Because each map is jointly simulated, pixels within any spatial region can be aggregated together to define a regional mean, and the 500 different versions of that mean across the set of maps provides a model of the uncertainty for that mean value. Simulation is constrained to the global limits of stable transmission.