Particle MCMC in action

In connection with our discussion of the Read Paper by Christophe Andrieu, Arnaud Doucet and Roman Holenstein, Particle Markov chain Monte Carlo, my PhD student Pierre Jacob went to the extent of making his Python code public and open so that anyone can test the central algorithm by simply modifying the model on which it is tested. The code is available on Google.code, another interesting feature of Google for sharing information, and is called py-mmh for Python-based marginal Metropolis-Hastings. The site contains a wiki with items of information on how to install the program and more importantly how to modify it towards your own model.