Fig. 4
From: Data-driven prediction of COVID-19 cases in Germany for decision making
Merged Approaches for the example of Germany. The two approaches differ in their data handling strategies for considering reporting delays: Approach 1 (panel a) simply ignores the two latest data points. Approach 2, in contrast, uses estimated correction factors on the latest data points (panel b). The result of the merging (panel c) indicates that both approaches describe the data well, but make differing predictions. Therefore the resulting uncertainty is bigger than the individual uncertainties. In general, this procedure generalizes to more different approaches