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

  • Authors:

  • Alessandra Iacobucci;Jean-Michel Marin;Christian Robert

  • Affiliations:

  • CEREMADE, Université Paris Dauphine, France;Institut de Mathématiques et Modélisation de Montpellier (UMR CNRS 5149), Université Montpellier 2, France and CREST, INSEE, France;CEREMADE, Université Paris Dauphine, France and CREST, INSEE, France

  • Venue:
  • Computational Statistics & Data Analysis
  • Year:
  • 2010

Quantified Score

Hi-index 0.03

Visualization

Abstract

Population Monte Carlo has been introduced as a sequential importance sampling technique to overcome poor fit of the importance function. The performance of the original Population Monte Carlo algorithm is compared 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 a large simulation experiment on posterior distributions of mean mixtures of distributions.