Generate diagnostic plots for a fitted Bayesian multivariate probit model

Generate diagnostic plots for a fitted Bayesian multivariate probit model

Usage

plot_probit_diagnostics(
  fit_obj,
  output_dir,
  experiment_id,
  pathogen,
  max_params = 30L
)

Arguments

fit_obj

Named list returned by fit_bayesian_multivariate_probit().

output_dir

Directory where the PDF will be saved.

experiment_id

Character. Experiment identifier (used in filename and titles).

pathogen

Character. Pathogen name (used in plot titles).

max_params

Integer. Maximum number of parameters shown in trace/rank plots. Defaults to 30. Includes lp__, all tau_*, and controlled subsets of beta, random-effect terms, random-effect correlations, and residual-correlation terms when available.

Value

Invisibly returns the path to the saved PDF, or NULL if skipped.