Post_Prob_R: Bayesian posterior probability, given an observed difference...

In IDDI-BE/PhII_Bayes: Bayesian tools for PhII clinical trial design

Description

This function calculates the Bayesian posterior probability P(Delta>Dcut|n,x) with Delta=Delta_exp-Delta_ctrl, with Delta_exp~beta(beta_par_exp[1]+x,beta_par_exp[2]+n_exp-x_exp) with prior beta distribution beta_par_exp, and Delta_ctrl~beta(beta_par_ctrl[1]+x,beta_par_ctrl[2]+n_ctrl-x_ctrl)

Usage

Post_Prob_R(
  p_exp,
  p_ctrl,
  n_exp,
  n_ctrl,
  distrisize,
  Dcut,
  beta_par_exp,
  beta_par_ctrl
)

Arguments

p_exp	observed proportion experimental arm
p_ctrl	observed proportion control arm
n_exp	number of patients in experimental arm (scalar)
n_ctrl	number of patients in control arm (scalar)
distrisize	Size of sampled distributions (the larger, the better)
Dcut	True difference between two proportions (can be a vector)
beta_par_exp	two shape parameters c(alpha,beta) for prior beta distribution experimental arm (scalar)
beta_par_ctrl	two shape parameters c(alpha,beta) for prior beta distribution control arm (scalar)

Examples

Post_Prob_R(p_exp=0.475,p_ctrl=0.475-5/50,n_exp=50,n_ctrl=50,distrisize=10^3,
Dcut=c(0,0.05,0.075,0.1),beta_par_exp=c(1,1),beta_par_ctrl=c(1,1))

IDDI-BE/PhII_Bayes documentation built on May 19, 2021, 3:04 p.m.