For example, for a threshold probability of 15%, a predictor being equally sensitivity and specific to the event, and evaluated at timepoint 1, the true net benefit was 0.0185 and the mean net benefit over 2000 replications was 0.0186.
By setting, for instance, a threshold probability of 0.9, we find that, for the higher SER, the UEP MSEs are lower than 0.009 whereas for EEP they are lower than 0.05.
For a threshold probability of 15%, the uncorrected estimate was over-optimistic for all scenarios; all correction methods gave an estimate lower than the best estimate of net benefit; repeated 10-fold cross-validation had the least bias for all but the scenario with 100 events, where the bootstrap estimate had slightly lower bias (-0.0001 vs -0.0005).
For a threshold probability of 40%% the net benefit obtained is 6.31%%.
We determined the coverage of the method described above, for a given time t and for a threshold probabilities of 15, 30, and 60%.
However, it is unlikely that a physician would consider an unnecessary biopsy to be worse than missing a cancer, so the threshold probability for biopsy must be less than 50%.
The threshold probability for a sequence match was set at 10−5.
Determination of the threshold probability for predicted presence using TSS resulted in a mean proportion of correctly classified training observations of 0.925 ± 0.016.
The threshold probability for 90th percentile (10% probability for death occurrence) during summer season was 65 death counts, for 95th (5%) was 68 deaths and 99th percentile of daily mortality (1%) would happened if mortality surpasses 72 dead.
The threshold probability for function annotation is 0.470, in which we expect 50% precision for the protein function prediction from the cross validation experiment in Section 3.1.
