Premier séminaire de statistique – H09
Le jeudi 22 janvier 2009, à 13 h 30,
à la salle 1240 3870 (changement de salle) du pavillon Alexandre-Vachon
Learn from Thy Neighbour: Parallel-Chain Adaptive MCMC
Department of Statistics, University of Toronto.
A considerable amount of effort has been recently invested in developing a comprehensive theory for adaptive MCMC. In comparison, there are fewer adaptive algorithms designed for practical situations. I will review some of the theoretical approaches used for proving convergence of non-Markovian adaptation schemes and will discuss scenarios for which the original adaptive Random-Walk Metropolis is unsuitable. Alternative adaptive schemes involving inter-chain and regional adaptation are discussed. The theory is illustrated with simulated and real examples.
This is joint work with Antonio Fabio di Narzo (Statistics, Bologna), Jeffrey Rosenthal (Statistics, Toronto) and Chao Yang (Statistics, Toronto).