Table 1. Model selection of transmission scenarios.
Model | Description | WAIC | ΔWAIC | WBIC | ΔWBIC |
---|---|---|---|---|---|
A | Reactivation and re-infection | 22156.2 | 13.0 | 22211.3 | 13.7 |
B | No re-infection | 22155.5 | 12.3 | 22209.4 | 11.8 |
C | No reactivation | 22363.6 | 220.4 | 22396.9 | 199.3 |
D | Reactivation/re-infection not infectious | 22215.3 | 72.1 | 22247.9 | 36.6 |
E | Multiple reactivations/re-infections | 22144.0 | 0.7 | 22197.6 | 0 |
F | Multiple reactivations and no re-infection | 22143.2 | 0 | 22198.4 | 0.8 |
G | Multiple re-infections and no reactivation | 22364.4 | 221.2 | 22396.8 | 199.2 |
For each of seven model scenarios we report the WAIC, a measure of predictive performance, and WBIC, a measure for the most likely model generating the data. Also shown are the WAIC and WBIC differences with the best fitting model. Models E-G contain the possibility of multiple reactivation/re-infection events in persons with increased antibody concentrations (the B compartment; cf. Fig 2).