Fig. 4From: Space-time modelling of monthly malaria incidence for seasonal associated drivers and early epidemic detection in Southern EthiopiaThe figures show the outcomes of the Monte Carlo methods applied to test the temporal independence hypothesis (upper panels) and the data compatibility hypothesis (bottom panels) for each district. The 95% confidence region is represented by the shaded areas for each hypothesis. The solid lines represent the empirical variogram for each time bin. Using P. falciparum count data from Southern Ethiopia, the theoretical variograms derived using the least squares (solid lines) and maximum likelihood (dashed lines) approaches are displayed in the lower panelBack to article page