Welcome to the MCMC Preprint Service
This page contains lists to MCMC-related sites including those holding
software, codes and applets that may be useful
or of interest to people working in the MCMC area.
- McDiag Home Page
This is an extremely useful site providing all sorts of useful information about
recent and current research on MCMC Convergence Diagnosis.
- Perfect Simulation
Home Page This is another useful site, keeping track of current and recent research
on perfect simulation.
Monte Carlo Methods. This site provides
material on sequential Monte Carlo methods focussing
attention on particle filtering. There is also a
US Mirror of the
- The Bugs Software Package.
This is a powerful statistical package for performing MCMC simulations. There is
also a US mirror.
HYDRA is an open-source, platform-neutral library for performing Markov
Chain Monte Carlo. It implements the logic of standard MCMC samplers within
a framework designed to be easy to use and to extend while allowing integration
with other software tools.
It provides methods for implementing MCMC samplers using Metropolis,
Metropolis-Hastings, Gibbs methods as well as classes implementing
several unique adaptive and multiple chain/parallel MCMC methods.
- Bayesian Output Analysis
An excellent S-PLUS/R program for carrying out convergence diagnostics and statistical and
graphical analysis of Monte Carlo sampling output. It can be used as an
output processor for the BUGS software or for any other program which produces sampling output.
Jeff Rosenthal's Java Applets.
This a page containing all sorts of Java applets including some illustrative
MCMC samplers, such as the
independence sampler with exponential target distribution,
Metropolis sampler with exponential target distribution
and uniform proposal distributions, and
the slice sampler.
The latest addition to this set of applets is one to simulate a
Coupling from the Past
Dead leaves simulation.
This is a very pretty simulation which discusses the ideas of forward and backward
simulation and provides a practical illustration of a perfect simulation.
C-Library for Gaussian Random Fields.
Provides a C-library for fast and exact sampling of Gaussian Markov random fields
(works for both lattices and graphs) and is intended for block-sampling in MCMC.
Flexible Bayesian Modeling and MCMC Software. Lets you try
out a variety of Markov chain methods on distributions specified
by simple formulas (including a BUGS-like notation), and also
on more complex models based on neural networks, Gaussian processes,
- Laird's Applet Page.
This page provides links to a variety of java applets illustrating different MCMC
Administrator: Steve Brooks
Statistical Laboratory, University of Cambridge