classification_summary_cv {bayesrules}R Documentation

Cross-Validated Posterior Classification Summaries

Description

Given a set of observed data including a binary response variable y and an rstanreg model of y, this function returns cross validated estimates of the model's posterior classification quality: sensitivity, specificity, and overall accuracy. For hierarchical models of class lmerMod, the folds are comprised by collections of groups, not individual observations.

Usage

classification_summary_cv(model, data, group, k, cutoff = 0.5)

Arguments

model

an rstanreg model object with binary y

data

data frame including the variables in the model, both response y (0 or 1) and predictors x

group

a character string representing the name of the factor grouping variable, ie. random effect (only used for hierarchical models)

k

the number of folds to use for cross validation

cutoff

probability cutoff to classify a new case as positive

Value

a list

Examples

x <- rnorm(20)
z <- 3*x
prob <- 1/(1+exp(-z))
y <- rbinom(20, 1, prob)
example_data <- data.frame(x = x, y = y)
example_model <- rstanarm::stan_glm(y ~ x, data = example_data, family = binomial)
classification_summary_cv(model = example_model, data = example_data, k = 2, cutoff = 0.5)                   

[Package bayesrules version 0.0.2 Index]