In many challenging statistical inference problems of the 21st century, the existing likelihood-based methods, such as Markov chain Monte Carlo methods, the EM algorithm and Sequential Monte Carlo methods are limited and cannot address the high dimensionality and massive data sets.
However, breakthroughs in computational and statistical approaches hold promise to address intractable likelihood problems, including:
- pseudo-marginal and particle MCMC
- likelihood-free methods such as Approximate Bayesian Computation
- composite and pseudo-likelihoods, new simulation methods for hitherto intractable stochastic models
- adaptive Monte Carlo methods
These advances coupled with developments in multi-core computational technologies such as GPUs, have enormous potential for extending likelihood methods to meet the most difficult challenges of modern scientific questions.
This workshop will bring together leaders in the field of computational statistics and scientific computing to discuss their latest ideas and present methods for addressing complex inference problems.