Format

Send to

Choose Destination
Stat Med. 2015 Dec 10;34(28):3696-713. doi: 10.1002/sim.6596. Epub 2015 Aug 4.

On computer-intensive simulation and estimation methods for rare-event analysis in epidemic models.

Author information

1
Institut Telecom LTCI UMR Telecom ParisTech/CNRS No. 5141, F-75634, Paris, France.
2
INSERM, IAME, UMR 1137, Paris, F-75018, France.
3
IAME, UMR 1137, Univ Paris Diderot, Sorbonne Paris Cité, F-75018, Paris, France.
4
Université Pierre et Marie Curie LPMA UMR CNRS, No. 7599, Paris, France.
5
Laboratoire P. Painlevé UFR de Mathématiques UMR CNRS 8524, Université des Sciences et Technologies Lille 1, Villeneuve d'Ascq Cedex, F-59955, France.

Abstract

This article focuses, in the context of epidemic models, on rare events that may possibly correspond to crisis situations from the perspective of public health. In general, no close analytic form for their occurrence probabilities is available, and crude Monte Carlo procedures fail. We show how recent intensive computer simulation techniques, such as interacting branching particle methods, can be used for estimation purposes, as well as for generating model paths that correspond to realizations of such events. Applications of these simulation-based methods to several epidemic models fitted from real datasets are also considered and discussed thoroughly.

KEYWORDS:

Monte Carlo simulation; genetic models; importance sampling; interacting branching particle system; multilevel splitting; rare-event analysis; stochastic epidemic model

PMID:
26242476
DOI:
10.1002/sim.6596
[Indexed for MEDLINE]

Supplemental Content

Loading ...
Support Center