Binary Calibration Uncertainty — binary_calibration_uncertainty • gp3bayes

Equal-width bins are defined from posterior-mean predicted probabilities.

Usage

binary_calibration_uncertainty(
  fit,
  newdata = NULL,
  bins = 10L,
  include_group_effects = FALSE,
  ndraws = 1000L,
  probs = c(0.025, 0.5, 0.975)
)

Arguments

fit: A fitted binary gp3bayes_fit.
newdata: Optional data containing observed outcomes.
bins: Number of equal-width probability bins.
include_group_effects: Whether fitted group effects are included.
ndraws: Expected-probability posterior draws.
probs: Three interval probabilities.

Value

A gp3bayes_binary_calibration_uncertainty.