Thumbnail - No thumbnail

On variance stabilisation in population Monte Carlo by double Rao-Blackwellisation

Robert, Christian P.; Marin, Jean-Michel; Iacobucci, Alessandra (2010), On variance stabilisation in population Monte Carlo by double Rao-Blackwellisation, Computational Statistics and Data Analysis, 54, 3, p. 698-710. http://dx.doi.org/10.1016/j.csda.2008.09.020

Type
Article accepté pour publication ou publié
External document link
http://hal.inria.fr/inria-00260141/en/
Date
2010
Journal name
Computational Statistics and Data Analysis
Volume
54
Number
3
Publisher
Elsevier
Pages
698-710
Publication identifier
Metadata
Show full item record
Author(s)
Robert, Christian P.
Marin, Jean-Michel cc
Iacobucci, Alessandra cc
Abstract (EN)
Population Monte Carlo has been introduced as a sequential importance sampling technique to overcome poor fit of the importance function. In this paper, we compare the performances of the original Population Monte Carlo algorithm with a modified version that eliminates the influence of the transition particle via a double Rao-Blackwellisation. This modification is shown to improve the exploration of the modes through an large simulation experiment on posterior distributions of mean mixtures of distributions.
Subjects / Keywords
Importance sampling mixture; Adaptive Monte Carlo; Population Monte Carlo; Multimodality; Mixture of distributions; Random walk
JEL
C15 - Statistical Simulation Methods: General