|
Activity Number:
|
455
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #304036 |
|
Title:
|
Adaptive Multiple Importance Sampling
|
|
Author(s):
|
Jean-Michel Marin*+ and Jean-Marie Cornuet and Antonietta Mira and Christian P. Robert
|
|
Companies:
|
University Montpellier II and Imperial College, London and University of Insubria and Université Paris Dauphine
|
|
Address:
|
Department of Mathematics, , ,
|
|
Keywords:
|
adaptive importance sampling ; sequential Monte Carlo ; t-distribution ; population Genetics
|
|
Abstract:
|
The Adaptive Multiple Importance Sampling algorithm is aimed at an optimal recycling of past simulations in an iterated importance sampling scheme. The difference with earlier adaptive importance sampling implementations like Population Monte Carlo is that the importance weights of all simulated values, past as well as present, are recomputed at each iteration, following the technique of the deterministic multiple mixture estimator of Owen and Zhou (JASA, 2000). Although the convergence properties of the algorithm cannot be fully investigated, we demonstrate through a population genetics example that the improvement brought by this technique is significant.
|
