Internal function for the Mixed-Effects Model Calibration Curve (MIXC)

Estimates the calibration curve using a logistic generalized linear mixed model.

Usage

MIXC(
  data = NULL,
  p,
  y,
  cluster,
  grid,
  method = c("slope", "intercept"),
  plot = TRUE,
  cluster_curves = FALSE,
  nsims_pi = 10000,
  CI = TRUE,
  CI_method = c("naive", "delta"),
  cl.level = 0.95
)

Arguments

data: optional data frame containing the variables p, y, and cluster. If supplied, variable names should be given without quotation marks.
p: predicted probabilities (numeric vector) or name of the column in data.
y: binary outcome variable or the name of the column in data.
cluster: Cluster identifier (factor, character, or integer) or name of the column in data.
grid: the grid for the calibration curve evaluation
method: character, type of mixed-effects model: "intercept" (random intercept) or "slope" (random slope). Default is "slope".
plot: logical, indicating whether to generate a calibration plot. Default is TRUE.
cluster_curves: logical, whether to include cluster-specific curves in the plot. Default is FALSE.
nsims_pi: integer, number of simulations for prediction intervals. Default is 10000.
CI: logical, whether to calculate confidence intervals. Default is TRUE.
CI_method: character, method for computing the confidence intervals of the observed proportions. If "delta", the delta method is applied. Conversely, when CI_method == "naive", no correction is applied. Default is "naive".
cl.level: the confidence level for the calculation of the confidence interval. Default is 0.95.

Value

A list containing:

model: The fitted mixed-effects model object
cluster_data: Data frame with calibration data for each cluster
plot_data: Data frame with calibration data for the average cluster
observed_data: Data frame with calibration data for individual observations
plot: A ggplot2 object if plot = TRUE, otherwise NULL

Details

This function estimates the calibration curves using a logistic generalized linear mixed model.