Table 2 Mean squared errors of the investigated methods, using the parametrization with the lowest MSE for each scenario (data shown as mean ± standard deviation)
From: Comparing methods to predict baseline mortality for excess mortality calculations
Method | Constant | Linear trend | Quadratic trend | Non-monotone |
|---|---|---|---|---|
AI | 1182.2 ± 1273.6 | 73.3 ± 77.4 | 388.2 ± 553.5 | 8380.2 ± 14478 |
Average | 969.0 ± 596.0 | 961.3 ± 753.1 | 929.6 ± 1039.6 | 8013.8 ± 4529.7 |
Linear | 1150.0 ± 1213.7 | 70.3 ± 73.1 | 365.5 ± 524.0 | 10,702.6 ± 11901.7 |
WHO | 1418.4 ± 2075.1 | 89.1 ± 123.8 | 552.5 ± 824.5 | 8067.2 ± 9866.4 |