optCutoff: Compute the Optimal Cutoff for Binary Classification

In MKmisc: Miscellaneous Functions from M. Kohl

Compute the Optimal Cutoff for Binary Classification

Description

The function computes the optimal cutoff for various performance weasures for binary classification.

Usage

optCutoff(pred, truth, namePos, perfMeasure = "Youden's J statistic",
          max = TRUE, parallel = FALSE, ncores, delta = 0.01)

Arguments

pred	numeric values that shall be used for classification; e.g. probabilities to belong to the positive group.
truth	true grouping vector or factor.
namePos	value representing the positive group.
perfMeasure	a performance measure computed by function perfMeasure.
max	logical value. Whether to maximize or minimize the performacne measure.
parallel	logical value. If TRUE packages foreach and doParallel are used to parallelize the computations.
ncores	integer value, number of cores that shall be used to parallelize the computations.
delta	numeric value for setting up grid for optimization; start is minimum of pred-delta, end is maximum of pred+delta.

Details

The function is ablte to compute the optimal cutoff for various performance measures, all performance measures that are implemented in function perfMeasures.

Value

Optimal cutoff and value of the optimized performance measure based on a simple grid search.

Author(s)

Matthias Kohl [e-mail]

Examples

## example from dataset infert
fit <- glm(case ~ spontaneous+induced, data = infert, family = binomial())
pred <- predict(fit, type = "response")
optCutoff(pred, truth = infert$case, namePos = 1)

MKmisc documentation built on Nov. 20, 2022, 1:05 a.m.