Thumbnail - No thumbnail

Efficient learning in ABC algorithms

Obert, Christian P. R; Pudlo, Pierre; Marin, Jean-Michel; Cornuet, Jean-Marie; Sedki, Mohammed (2012), Efficient learning in ABC algorithms. https://basepub.dauphine.fr/handle/123456789/10474

Type
Document de travail / Working paper
External document link
http://fr.arxiv.org/abs/1210.1388
Date
2012
Publisher
Université Paris-Dauphine
Published in
Paris
Pages
24
Metadata
Show full item record
Author(s)
Obert, Christian P. R

Pudlo, Pierre

Marin, Jean-Michel cc

Cornuet, Jean-Marie

Sedki, Mohammed
Abstract (EN)
Approximate Bayesian Computation has been successfully used in population genetics models to bypass the calculation of the likelihood. These algorithms provide an accurate estimator by comparing the observed dataset to a sample of datasets simulated from the model. Although parallelization is easily achieved, computation times for assuring a suitable approximation quality of the posterior distribution are still long. To alleviate this issue, we propose a sequential algorithm adapted from Del Moral et al. (2012) which runs twice as fast as traditional ABC algorithms. Its parameters are calibrated to minimize the number of simulations from the model.
Subjects / Keywords
population genetics; Approximate Bayesian Computation; Sequential Monte Carlo; likelihood-free sampler
JEL
C15 - Statistical Simulation Methods: General
C11 - Bayesian Analysis: General