Skip to contents

ROC related computations – internal function

Source: R/roc_with_ci.R
roc_with_ci.Rd

ROC related computations – internal function

Usage

roc_with_ci(
  probs,
  labels,
  positive,
  auc,
  ci_method = c("none", "delong", "bootstrap"),
  n_boot = 1000,
  alpha = 0.05
)

Arguments

probs

Vector of class probabilities or values of continuous predictor

labels

Vector, factor with two levels

positive

Character giving the level that corresponds to 'case'

auc

Numeric value of AUC

ci_method

Character from c("none", "delong", "bootstrap")

n_boot

Numeric value giving the number of bootstrap replicates (default: 1000)

alpha

Level of significance (default: 0.05)

Value

List with components:

roc_df

data frame for plotting ROC curve

roc_band

data frame for plotting confidence band of ROC

auc

auc

auc_ci

confint for auc

Details

ci_method = 'bootstrap' makes this function's result stochastic (resampled AUC and ROC confidence band). For reproducible results, call set.seed() immediately beforehand.