mcmc: Markov Chain Monte Carlo

Simulates continuous distributions of random vectors using Markov chain Monte Carlo (MCMC). Users specify the distribution by an R function that evaluates the log unnormalized density. Algorithms are random walk Metropolis algorithm (function metrop), simulated tempering (function temper), and morphometric random walk Metropolis (Johnson and Geyer, Annals of Statistics, 2012, function morph.metrop), which achieves geometric ergodicity by change of variable.

Version: 0.9-4
Depends: R (≥ 2.10.0)
Imports: stats
Suggests: xtable, Iso
Published: 2015-07-17
Author: Charles J. Geyer and Leif T. Johnson
Maintainer: Charles J. Geyer <charlie at>
License: MIT + file LICENSE
NeedsCompilation: yes
Materials: ChangeLog
In views: Bayesian
CRAN checks: mcmc results


Reference manual: mcmc.pdf
Vignettes: Bayes Factors via Serial Tempering
Debugging MCMC Code
MCMC Example
MCMC Morph Example
Package source: mcmc_0.9-4.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X Snow Leopard binaries: r-release: mcmc_0.9-4.tgz, r-oldrel: mcmc_0.9-3.tgz
OS X Mavericks binaries: r-release: mcmc_0.9-4.tgz
Old sources: mcmc archive

Reverse dependencies:

Reverse depends: ltbayes
Reverse imports: ReliabilityTheory, TBSSurvival
Reverse suggests: pse