Weighted multilevel Langevin simulation of invariant measures

DOI10.1214/17-AAP1364MaRDI QIDQ1634175zbMATH OpenOpenAlexWikidataFDO

Authors Gilles Pagès, Fabien Panloup

Publication date 17 December 2018

Published in The Annals of Applied Probability (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1607.00781, https://projecteuclid.org/euclid.aoap/1538985625

zbMATH Keywords

ergodic diffusionergodicityinvariant measureMonte CarlomultilevelPAC-BayesianRichardson-Romberg

Mathematics Subject Classification ID

Computational methods for ergodic theory (approximation of invariant measures, computation of Lyapunov exponents, entropy, etc.) (37M25) Diffusion processes (60J60) Monte Carlo methods (65C05)

Abstract: We investigate a weighted Multilevel Richardson-Romberg extrapolation for the ergodic approximation of invariant distributions of diffusions adapted from the one introduced in~[Lemaire-Pag`es, 2013] for regular Monte Carlo simulation. In a first result, we prove under weak confluence assumptions on the diffusion, that for any integer $Rge2$, the procedure allows us to attain a rate $n^{fracR2R+1}$ whereas the original algorithm convergence is at a weak rate $n^{1/3}$. Furthermore, this is achieved without any explosion of the asymptotic variance. In a second part, under stronger confluence assumptions and with the help of some second order expansions of the asymptotic error, we go deeper in the study by optimizing the choice of the parameters involved by the method. In particular, for a given $varepsilonextgreater0$, we exhibit some semi-explicit parameters for which the number of iterations of the Euler scheme required to attain a Mean-Squared Error lower than $varepsilo{n}^2$ is about $varepsilo{n}^{-2}log(varepsilo{n}^{-1})$. Finally, we numerically this Multilevel Langevin estimator on several examples including the simple one-dimensional Ornstein-Uhlenbeck process but also on a high dimensional diffusion motivated by a statistical problem. These examples confirm the theoretical efficiency of the method.

Recommendations

• Langevin-type models. I: Diffusions with given stationary distributions and their discretizations
• Multi-level Monte Carlo methods for the approximation of invariant measures of stochastic differential equations
• Multilevel Richardson-Romberg extrapolation
• Limit theorems for weighted and regular multilevel estimators
• Invariant measure of duplicated diffusions and application to Richardson-Romberg extrapolation

Cites work

• A decreasing step method for strongly oscillating stochastic models
• A mixed-step algorithm for the approximation of the stationary regime of a diffusion
• An Iterative Monte Carlo Scheme for Generating Lie Group-Valued Random Variables
• Antithetic multilevel Monte Carlo estimation for multi-dimensional SDEs without Lévy area simulation
• Approximation of the distribution of a stationary Markov process with application to option pricing
• Behavior of the Euler scheme with decreasing step in a degenerate situation
• Ergodic theorems. With a supplement by Antoine Brunel
• Exponential convergence of Langevin distributions and their discrete approximations
• scientific article; zbMATH DE number 3723610 (Why is no real title available?)
• scientific article; zbMATH DE number 976356 (Why is no real title available?)
• scientific article; zbMATH DE number 3895476 (Why is no real title available?)
• scientific article; zbMATH DE number 3227613 (Why is no real title available?)
• Invariant measure of duplicated diffusions and application to Richardson-Romberg extrapolation
• Multi-level stochastic approximation algorithms
• Multi-step Richardson-Romberg Extrapolation: Remarks on Variance Control and Complexity
• Multilevel Monte Carlo Path Simulation
• Multilevel Richardson-Romberg extrapolation
• On the functional central limit theorem and the law of the iterated logarithm for Markov processes
• On the Poisson equation and diffusion approximation. I
• Optimal Scaling of Discrete Approximations to Langevin Diffusions
• Recursive computation of the invariant distribution of a diffusion
• RECURSIVE COMPUTATION OF THE INVARIANT DISTRIBUTION OF A DIFFUSION: THE CASE OF A WEAKLY MEAN REVERTING DRIFT
• Recursive computation of the invariant measure of a stochastic differential equation driven by a Lévy process
• Second-order discretization schemes of stochastic differential systems for the computation of the invariant law
• Sparse regression learning by aggregation and Langevin Monte-Carlo

Cited in

(4)

• Variance reduction for additive functionals of Markov chains via martingale representations
• Discretization of the ergodic functional central limit theorem
• (Non)-penalized multilevel methods for non-uniformly log-concave distributions
• Invariant measure of duplicated diffusions and application to Richardson-Romberg extrapolation

This page was built for publication: Weighted multilevel Langevin simulation of invariant measures

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1634175)