A class for Gaussian Robust Mixture Prior models using RBesT for inference.
Details
This class represents a Gaussian Robust Mixture Prior model. It inherits from the Model class. Inference is performed using RBesT.
This class extends the Model_RBesT class and provides methods for initializing the model, updating priors, calculating posterior moments, and converting distributions to RBesT format.
Super classes
Model -> Model_RBesT -> GaussianRMP_RBesT
Public fields
wThe weight of the prior distribution.
vague_prior_meanThe mean of the vague prior distribution.
vague_prior_varianceThe variance of the vague prior distribution.
info_prior_meanThe mean of the informative prior distribution.
info_prior_varianceThe variance of the informative prior distribution.
wpostThe weight of the posterior distribution.
vague_posterior_meanThe mean of the vague posterior distribution.
info_posterior_meanThe mean of the informative posterior distribution.
vague_posterior_varianceThe variance of the vague posterior distribution.
info_posterior_varianceThe variance of the informative posterior distribution.
methodMethod name
Methods
Inherited methods
Model$calibrate_for_design()Model$check_data()Model$create()Model$estimate_bayesian_operating_characteristics()Model$estimate_frequentist_operating_characteristics()Model$hypothesis_space_transformation()Model$inference()Model$inference_cache_scope()Model$plot_pdfs()Model$plot_posterior_pdf()Model$plot_prior_pdf()Model$posterior_beta_mixture()Model$posterior_ess()Model$posterior_quantile()Model$print_model_summary()Model$prior_elir_ess()Model$simulation_for_given_treatment_effect()Model$test_decision()Model_RBesT$credible_interval()Model_RBesT$posterior_cdf()Model_RBesT$posterior_mean()Model_RBesT$posterior_median()Model_RBesT$posterior_pdf()Model_RBesT$posterior_variance()Model_RBesT$prior_cdf()Model_RBesT$prior_pdf()Model_RBesT$sample_posterior()Model_RBesT$sample_prior()Model_RBesT$vectorised_replicate_inference()
GaussianRMP_RBesT$new()
Initialize a new GaussianRMP_RBesT object.
Usage
GaussianRMP_RBesT$new(prior)GaussianRMP_RBesT$empirical_bayes_update()
Update the vague prior variance based on empirical Bayes approach.
GaussianRMP_RBesT$vectorised_prior_components()
Prior mixture components for each replicate.
Under empirical Bayes the vague component's variance is re-derived from
each replicate, matching empirical_bayes_update(), so the prior varies
by row. Otherwise the same two components serve every replicate.
A degenerate weight collapses the mixture to a single component, exactly as the scalar path does to work around an RBesT ELIR bug.
GaussianRMP_RBesT$vectorised_posterior_parameters()
Posterior parameters reported by the vectorised path.
The scalar path records the posterior weight on the informative component, which is 0 or 1 when the mixture has collapsed.
Arguments
posteriorPosterior mixture from
normal_mixture_posterior().
GaussianRMP_RBesT$summary_rows()
Rows of the model summary, with the prior weight and the moments of the
two posterior components. The posterior weight is among the
posterior_parameters rows.