codadiags-package {codadiags}R Documentation

Markov chain Monte Carlo burn-in based on "bridge" statistics.

Description

This package implements many burn-in procedures based on iterative transient statistic tests. The tests are calibrated on simple AR1 MCMC process, tuned on particle simulation Monte Carlo codes. It should be viewed as a non asymptotic declination of heidel.diag burnin from coda package.

Details

Package: codadiags
Type: Package
Version: 1.0
Date: 2013-09-17
License: GPL-3
Depends: coda

Author(s)

Yann Richet, Xavier Bay, Olivier Jacquet, Alexis Jinaphanh Maintainer: Yann Richet <yann.richet@irsn.fr>

References

Y. Richet, PhD: "Suppression of the initial transient in Monte Carlo criticality simulations", University of St Etienne, France, 2009.

Examples

require(codadiags)
set.seed(123)
x = AR1()
print(bridgestat.diag(x))
y = add.transient(x)
print(bridgestat.diag(y,trunc=10))

[Package codadiags version 1.0 Index]