aucCI: Calculate CI for AUC

View source: R/roc_glm_ci.R

aucCIR Documentation

Calculate CI for AUC

Description

This function calculates a CI for the AUC based on the approach proposed by DeLong.

Usage

aucCI(
  connections,
  truth_name,
  pred_name,
  roc_glm,
  alpha = 0.05,
  epsilon = 0.2,
  delta = 0.2,
  seed_object = NULL
)

Arguments

connections

('DSI::connection') Connection to an OPAL server.

truth_name

('character(1L)') Character containing the name of the vector of 0-1-values encoded as integer or numeric.

pred_name

('character(1L)') Character containing the name of the vector of probabilities.

roc_glm

('list()') List containing the ROC-GLM parameter returned from 'dsROCGLM'.

alpha

('numeric(1L)') Significance level alpha (default is '0.05').

epsilon

('numeric(1L)') Privacy parameter for differential privacy (DP).

delta

('numeric(1L)') Probability of violating epsilon DP.

seed_object

('character(1L)') Name of an object which is used to add a seed based on an object.

Value

Numeric vector with two values containing the boundaries of the confidence interval.

Author(s)

Daniel S.


difuture-lmu/dsROCGLM documentation built on March 24, 2024, 1:07 p.m.