Statistical inference for rough Burgers-type stochastic partial differential equations
Abstract
We consider the Burgers-type stochastic PDE du = θ ∂x2 u dt + f(u) dt + g(u) ∂x u dt + σ dW(t), driven by space--time white noise on the one-dimensional torus, where θ>0 is the diffusivity parameter and σ>0 the noise intensity. The transport term g(u) ∂x u admits no classical interpretation, and we work with the mild solutions defined via rough paths following~\cite{Hairer2011}, for which the solution decomposes as u=ψ+R with ψ a stationary centred Gaussian process solving a damped stochastic heat equation and R a remainder. We first show that, away from the initial time, R is Lipschitz in space and Hölder in time of order strictly greater than 1/2, hence strictly smoother than ψ in both variables. We then compute exactly the spatial quadratic variation and the temporal quartic variation of ψ, almost surely and not merely in L2(Ω), and prove that the corresponding variations of u are unaffected by the remainder. Combined with an exact time-scaling identity, this separation of regularity scales yields strongly consistent pathwise estimators of θ. To our knowledge this is the first inference procedure for an SPDE whose solution is defined only pathwise, as a measurable function of a rough path, rather than as an Itô process.
Related
- Inference for stochastic differential equations driven by weighted sub-fractional Brownian motion using neural networks and the Euler approximation
- A Priori Estimates for Singular Fractional Stochastic Burgers Equations
- Averaging principles for nonautonomous multiscale stochastic Burgers equations with reflection
- Mean field singular stochastic partial differential equations
- High-order fractional linear differential equations do not describe stochastic processes with finite state space
BibTeX
@misc{tudor2026statistical,
title = {{Statistical inference for rough Burgers-type stochastic partial differential equations}},
author = {Simon Chony Acosta Christian Olivera Ciprian Tudor},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.13673},
eprint = {2610.13673},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.13673},
}
RIS
TY - UNPB
TI - Statistical inference for rough Burgers-type stochastic partial differential equations
AU - Simon Chony Acosta Christian Olivera Ciprian Tudor
PY - 2026
DA - 2026/10/09
PB - arXiv
DO - 10.48550/arxiv.2610.13673
UR - https://arxiv.org/abs/2610.13673
AB - We consider the Burgers-type stochastic PDE du = θ ∂_x^2 u dt + f(u) dt + g(u) ∂_x u dt + σ dW(t), driven by space--time white noise on the one-dimensional torus, where θ>0 is the diffusivity parameter and σ>0 the noise intensity. The transport term g(u) ∂_x u admits no classical interpretation, and we work with the mild solutions defined via rough paths following~\cite{Hairer2011}, for which the solution decomposes as u=ψ+R with ψ a stationary centred Gaussian process solving a damped stochastic heat equation and R a remainder. We first show that, away from the initial time, R is Lipschitz in space and Hölder in time of order strictly greater than 1/2, hence strictly smoother than ψ in both variables. We then compute exactly the spatial quadratic variation and the temporal quartic variation of ψ, almost surely and not merely in L^2(Ω), and prove that the corresponding variations of u are unaffected by the remainder. Combined with an exact time-scaling identity, this separation of regularity scales yields strongly consistent pathwise estimators of θ. To our knowledge this is the first inference procedure for an SPDE whose solution is defined only pathwise, as a measurable function of a rough path, rather than as an Itô process.
ER -