Adaptive Importance Sampling in General Mixture Classes

Cappé, Olivier; Douc, Randal; Guillin, Arnaud; Marin, Jean-Michel; Robert, Christian P.

Cappé, Olivier; Douc, Randal; Guillin, Arnaud; Marin, Jean-Michel; Robert, Christian P. (2008), Adaptive Importance Sampling in General Mixture Classes, Statistics and Computing, 18, 4, p. 447-459. http://dx.doi.org/10.1007/s11222-008-9059-x

Type

Article accepté pour publication ou publié

External document link

http://hal.archives-ouvertes.fr/hal-00180669/en/

Date

2008

Journal name

Statistics and Computing

Volume

Number

Publisher

Springer

Pages

447-459

Abstract (EN)

In this paper, we propose an adaptive algorithm that iteratively updates both the weights and component parameters of a mixture importance sampling density so as to optimise the importance sampling performances, as measured by an entropy criterion. The method is shown to be applicable to a wide class of importance sampling densities, which includes in particular mixtures of multivariate Student t distributions. The performances of the proposed scheme are studied on both artificial and real examples, highlighting in particular the benefit of a novel Rao-Blackwellisation device which can be easily incorporated in the updating scheme.

Subjects / Keywords

Importance sampling; Adaptive Monte Carlo; Mixture model; Entropy; Kullback-Leibler divergence; EM algorithm; Population Monte Carlo