Run Predictive Check (Prior or Posterior) — run_predictive

Simulates responses from the specified Nimble model using either priors or a provided posterior distribution matrix. Generates corresponding diagnostic plots.

Usage

run_predictive_check(
  config,
  obs_X,
  posterior_samples = NULL,
  n_sim = 50,
  prefix = "Untitled",
  title = "PPC"
)

Arguments

config: Model configuration list from build_model_config.
obs_X: Numeric matrix of observed data to compare against.
posterior_samples: Optional mcmc.list or matrix of posterior samples. If NULL, simulates priors.
n_sim: Numeric. Number of simulations to draw. Default is 50.
prefix: Character or NULL. Prefix for the output plot filename. If NULL, plots render inline instead of saving to PDF.
title: Character. String used for labeling (e.g., "PriorPPC").

Value

A list containing the simulated and observed statistics.