fake (version 1.4.0)

ExpectedConcordance: Expected concordance statistic

Description

Computes the expected concordance statistic given true probabilities of event. In logistic regression, the concordance statistic is equal to the area under the Receiver Operating Characteristic (ROC) curve and estimates the probability that an individual who experienced the event (\(Y_i=1\)) had a higher probability of event than an individual who did not experience the event (\(Y_i=0\)).

Usage

ExpectedConcordance(probabilities)

Value

The expected concordance statistic.

Arguments

probabilities

vector of probabilities of event.

See Also

Concordance

Examples

# Simulation of probabilities
set.seed(1)
proba <- runif(n = 1000)

# Expected concordance
ExpectedConcordance(proba)

# Simulation of binary outcome
ydata <- rbinom(n = length(proba), size = 1, prob = proba)

# Observed concordance in simulated data
Concordance(observed = ydata, predicted = proba)