Bayesian numerical inference for Markovian models -- Application to tropical forest dynamics

Abstract : Bayesian modelling is fluently employed to assess natural ressources. It is associated with Monte Carlo Markov Chains (MCMC) to get an approximation of the distribution law of interest. Hence in such situations it is important to be able to propose N independent realiza- tions of this distribution law. We propose a strategy for making N parallel Monte Carlo Markov Chains interact in order to get an approximation of an independent N -sample of a given target law. In this method each individual chain proposes candidates for all other chains. We prove that the set of interacting chains is itself a MCMC method for the product of N target measures. Compared to independent parallel chains this method is more time consuming, but we show through example that it possesses many advantages. This approach will be applied to a forest dynamic model.