Calculates the standard error of MeasureSurvLogloss.

The standard error of the Logloss, L, is approximated via, $$se(L) = sd(L)/\sqrt{N}$$ where \(N\) are the number of observations in the test set, and \(sd\) is the standard deviation.

Censored observations in the test set are ignored.

Format

R6::R6Class() inheriting from MeasureSurvLogloss/MeasureSurv.

Construction

MeasureSurvLoglossSE$new(eps = 1e-15)
mlr_measures$get("surv.loglossSE")
msr("surv.loglossSE")
  • eps :: numeric(1)
    Very small number to set zero-valued predicted probabilities to, in order to prevent errors in log(0) calculation.

Meta Information

  • Type: "surv"

  • Range: \([0, \infty)\)

  • Minimize: TRUE

  • Required prediction: distr

Fields

See MeasureSurv, as well as all variables passed to the constructor.

See also