PPC_group_distribution: Posterior Predictive Check for Stan model

In ghurault/HuraultMisc: Guillem Hurault Functions' Library

Posterior Predictive Check for Stan model

Description

Plot the distribution density of parameters within a same group from a single/multiple draw of the posterior distribution. In the case of a hierarchical model, we might look at the distribution of patient parameter and compare it to the prior for the population distribution.

Usage

PPC_group_distribution(obj, parName = "", nDraws = 1)

Arguments

obj	Matrix (rows: samples, cols: parameter) or Stanfit object.
parName	Name of the observation-dependent (e.g. patient-dependent) parameter to consider (optional when obj is a matrix).
nDraws	Number of draws to plot

Value

Ggplot of the distribution

References

A. Gelman, J. B. B. Carlin, H. S. S. Stern, and D. B. B. Rubin, Bayesian Data Analysis (Chapter 6), Third Edition, 2014.

Examples

X <- matrix(rnorm(1e3), ncol = 10)
PPC_group_distribution(X, "", 10)

ghurault/HuraultMisc documentation built on Aug. 1, 2022, 3:11 p.m.