Antithetic Multilevel Particle Filters

DOI10.1017/APR.2024.12MaRDI QIDQ6424698zbMATH OpenFDO

Authors Ajay Jasra, Mohamed Maama, Hernando C. Ombao

Publication date 29 January 2023

Abstract: In this paper we consider the filtering of partially observed multi-dimensional diffusion processes that are observed regularly at discrete times. This is a challenging problem which requires the use of advanced numerical schemes based upon time-discretization of the diffusion process and then the application of particle filters. Perhaps the state-of-the-art method for moderate dimensional problems is the multilevel particle filter of cite{mlpf}. This is a method that combines multilevel Monte Carlo and particle filters. The approach in that article is based intrinsically upon an Euler discretization method. We develop a new particle filter based upon the antithetic truncated Milstein scheme of cite{ml_anti}. We show that for a class of diffusion problems, for $epsilon>0$ given, that the cost to produce a mean square error (MSE) in estimation of the filter, of $mathcalO(epsilo{n}^2)$ is $mathcalO(epsilo{n}^{-2}log(epsilon{)}^2)$. In the case of multidimensional diffusions with non-constant diffusion coefficient, the method of cite{mlpf} has a cost of $mathcalO(epsilo{n}^{-2.5})$ to achieve the same MSE. We support our theory with numerical results in several examples.

This page was built for publication: Antithetic Multilevel Particle Filters

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6424698)