Open Access
Table 6.
Bayesian comparison of candidate model classes.
| Model | k | Ndata | fcand | min.  |
BIC | p(M)/p(M*) |
|---|---|---|---|---|---|---|
| PL1 + cPL2 + γext | 14 | 13 | 0.26% | 1.32 | 37.23 | 0.94 |
| cPL1 + cPL2 + γext | 15 | 13 | 0.26% | 0.54 | 39.01 | 0.38 |
| PL1 + PL2 + rSIS | 14 | 15 | 0.03% | 1.60 | 39.51 | 0.04 |
| PL1 + PL2 + rNIS | 15 | 15 | 1.7% | 0.12 | 40.80 | 1 |
| PL1 + PL2 + rSIS + γext | 16 | 15 | 0.4% | 0.13 | 43.46 | 0.07 |
| PL1 + PL2 + rNIS + γext | 17 | 15 | 12% | 0.03 | 46.07 | 0.53 |
Notes. k is the number of free parameters, Ndata is the number of data points, and fcand is the weighted fraction of candidate models in the MCMC chain. The relative probability p(M)/p(M*) is calculated via Eq. (4) and is then renormalized.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.