skip to content
 

Monte Carlo Inference for Complex Statistical Models

22nd April 2014 to 25th April 2014

Organisers: Rong Chen (Rutgers), Dan Crisan (Imperial College London), Arnaud Doucet (Oxford), Simon Godsill (Cambridge), Ajay Jasra (Singapore) and Sumeetpal Singh (Cambridge)

Workshop Theme

In recent years the field of Monte Carlo methods for scientific computation has undergone explosive growth largely driven by the demands of real statistical problems, particularly those involving complex high-dimensional models. The challenges of inference for these complex statistical models has led to the development of algorithms of increased technical sophistication, a recent example being the interaction of Markov Chain Monte Carlo (MCMC) and Sequential Monte Carlo (SMC) to give rise to a class of algorithms referred to "exact approximations" of ideal MCMC algorithms.

One of the goals of this programme is to foster interaction and collaborations between researchers at the more technical side of these recent advances and those who wish to apply and develop Monte Carlo for real, high-dimensional, inference. This workshop will initiate the program by bringing together speakers who are contributing to the advancement of Monte Carlo methods and the list of topics covered includes:

  1. Approximate Bayesian Computation (ABC): new applications, methodology and theory
  2. Sequential/Markov chain Monte Carlo for high dimensional computation

The workshop will also have an introductory element to it aimed at acquainting non-specialists with the subject area. It will be comprised of a series of tutorial-style talks and invited talks, covering both the core methodology/principles of ABC, SMC and MCMC and presentations of recent results on the theory and practice of Monte Carlo based inference.

University of Cambridge Research Councils UK
    Clay Mathematics Institute The Leverhulme Trust London Mathematical Society Microsoft Research NM Rothschild and Sons