Pathwise central limit theorem and moderate deviations via rough paths for SPDEs with multiplicative noise

TL;DR

This paper develops a pathwise framework using rough path theory to establish a central limit theorem and moderate deviation principle for SPDEs with multiplicative noise, providing new insights into their asymptotic behavior.

Contribution

It introduces a novel rough path-based approach for CLT and MDP in SPDEs, avoiding weak convergence methods and applying to various equations including Landau-Lifschitz-Gilbert and reaction-diffusion.

Findings

Pathwise CLT and MDP established for several SPDEs.

Optimal convergence speed quantified in the CLT.

Framework applicable to equations driven by linear Itô noise.

Abstract

We put forward a general framework for the study of a pathwise central limit theorem (CLT) and a moderate deviation principle (MDP) for stochastic partial differential equations perturbed with a small multiplicative linear noise by means of the theory of rough paths. The CLT can be interpreted as the convergence to a pathwise derivative of the It\^o-Lyons map. The result follows by applying a pathwise Malliavin-like calculus for rough paths and from compactness methods. The convergence in the CLT is quantified by an optimal speed of convergence. From the exponential equivalence principle and the knowledge of the speed of convergence, we can derive easily a MDP. In particular, we do not apply the weak convergence approach usually employed in this framework. We derive a pathwise CLT and a MDP for the stochastic Landau-Lifschitz-Gilbert equation in one dimension, for the heat equation and for a stochastic reaction-diffusion equation. As a further application, we derive a pathwise convergence to the CLT limit and a corresponding MDP for equations driven by linear It\^o noise.