Limit theorems for cloning algorithms

STOCHASTIC PROCESSES AND THEIR APPLICATIONS

卷:138

Article

Angeli, Letizia^1,2,3  Grosskinsky, Stefan^1,2,4  Johansen, Adam M.^1,2 

[1] Univ Warwick, Math Inst, Coventry, W Midlands, England

[2] Univ Warwick, Dept Stat, Coventry, W Midlands, England

[3] Heriot Watt Univ, Dept Math, Edinburgh, Midlothian, Scotland

[4] Delft Univ Technol, Dept Appl Math DIAM, Delft, Netherlands

关键词: Cloning algorithm;    Dynamic large deviations;    Interacting particle systems;    L-p convergence;    Feynman-Kac formulae;    Jump processes;   

DOI  :  10.1016/j.spa.2021.04.007

来源: Elsevier

【 摘 要 】

Large deviations for additive path functionals of stochastic processes have attracted significant research interest, in particular in the context of stochastic particle systems and statistical physics. Efficient numerical 'cloning' algorithms have been developed to estimate the scaled cumulant generating function, based on importance sampling via cloning of rare event trajectories. So far, attempts to study the convergence properties of these algorithms in continuous time have led to only partial results for particular cases. Adapting previous results from the literature of particle filters and sequential Monte Carlo methods, we establish a first comprehensive and fully rigorous approach to bound systematic and random errors of cloning algorithms in continuous time. To this end we develop a method to compare different algorithms for particular classes of observables, based on the martingale characterization of stochastic processes. Our results apply to a large class of jump processes on compact state space, and do not involve any time discretization in contrast to previous approaches. This provides a robust and rigorous framework that can also be used to evaluate and improve the efficiency of algorithms. (C) 2021 The Author(s). Published by Elsevier B.V.

【 授权许可】

Free