Fig. 1: Model fit metrics graphs based on the simulation study. | Nature Communications

Fig. 1: Model fit metrics graphs based on the simulation study.

From: Estimating small area population from health intervention campaign surveys and partially observed settlement data

Fig. 1

The simulation study provides an opportunity to evaluate the performance of the proposed method against the conventional approach. Across the model fit metrics used, the TSBHM method consistently provided better model fits than the BHM approach. A Root Mean Square Error (RMSE), B Correlation Coefficient (CC) between the observed and predicted counts, C Error bar plots with 95% credible intervals and, D Violin plots of the predicted national total population across various proportions of satellite observation coverages using both BHM (blue) and TSBHM (orange) methods, when population data are 100% observed.

Back to article page