sim_pred_gamma_pois: Simulate from a gamma-Poisson posterior predictive...

In bang: Bayesian Analysis, No Gibbs

Simulate from a gamma-Poisson posterior predictive distribution

Description

Simulates nrep draws from the posterior predictive distribution of the beta-binomial model described in hef. This function is called within hef when the argument nrep is supplied.

Usage

sim_pred_gamma_pois(theta_sim_vals, data, nrep)

Arguments

theta_sim_vals	A numeric matrix with nrow(data) columns. Each row of theta_sim_vals contains binomial success probabilities simulated from their posterior distribution.
data	A 2-column numeric matrix: the numbers of successes in column 1 and the corresponding numbers of trials in column 2.
nrep	A numeric scalar. The number of replications of the original dataset simulated from the posterior predictive distribution. If nrep is greater than nrow(theta_sim_vals) then nrep is set equal to nrow(theta_sim_vals).

Value

A numeric matrix with nrep columns. Each column contains a draw from the posterior predictive distribution of the number of successes.

Examples

pump_res <- hef(model = "gamma_pois", data = pump)
pump_sim_pred <- sim_pred_gamma_pois(pump_res$theta_sim_vals, pump, 50)

bang documentation built on Sept. 11, 2024, 6:16 p.m.