Skip to contents

Get Original or Shrunk MCMC Draws for Subsequent Analyses (Updated)

Source: R/get_bayescores.R
get_bayescores_draws.Rd

Extracts or generates posterior draws for efficacy inputs. This version incorporates advanced shrinkage methods, including a skeptical prior via whitening, and strategically applies them based on the primary utility endpoint and parameter correlation.

Usage

get_bayescores_draws(
  fit,
  shrinkage_method = "none",
  shrinkage_target = "primary",
  calibration_args = list()
)

Arguments

fit

The fitted model object from `fit_bayesian_cure_model`.

shrinkage_method

A character string specifying the shrinkage method. Options are "zwet", "sherry", "skeptical", or "none" (default).

shrinkage_target

A character string controlling shrinkage application. Options are "primary" or "both_if_uncorrelated".

calibration_args

An optional list to override default utility calibration, used to determine the primary endpoint.

Value

A list containing two named vectors of posterior samples:

  • tr_posterior_samples: Draws for the Time Ratio (already exponentiated).

  • cure_posterior_samples: Draws for the absolute difference in cure rates.