IPA = 1 - BS(model) / BS(null), the null being the Kaplan-Meier
marginal survival. It is exactly the scaled_brier already
returned by Brier, so this is a thin alias rather than a second
implementation. IPA is 1 for a perfect model, 0 for a model no better
than the marginal, and negative for one that is worse.
Usage
Survipa(time, event, predicted_survival, eval_time)
Arguments
- time
Observed event or censoring times.
- event
Event indicator, 1 = event, 0 = censored.
- predicted_survival
Predicted survival at eval_time.
- eval_time
Evaluation time.
Value
List with estimate (IPA), brier_score,
scaled_brier, eval_time, method.
References
Kattan, M. W. and Gerds, T. A. (2018). The index of
prediction accuracy: an intuitive measure useful for evaluating risk
prediction models. Diagnostic and Prognostic Research 2:7.
doi:10.1186/s41512-018-0029-2
Examples
Survipa(c(1, 2, 3, 4), c(1, 1, 0, 1), c(0.9, 0.7, 0.5, 0.3), 2.5)
#> $estimate
#> [1] -1.04
#>
#> $brier_score
#> [1] 0.51
#>
#> $scaled_brier
#> [1] -1.04
#>
#> $eval_time
#> [1] 2.5
#>
#> $method
#> [1] "IPA = 1 - BS_model/BS_null [Kattan & Gerds 2018]"
#>
