Strong convergence of some Magnus-type schemes for the finite element discretization of non-autonomous parabolic SPDEs driven by additive fractional Brownian motion and Poisson random measure

Aurelien Junior Noupelah,Jean Daniel Mukam,Antoine Tambue

2024-09-10

Abstract:The aim of this work is to provide the strong convergence results of numerical approximations of a general second order non-autonomous semilinear stochastic partial differential equation (SPDE) driven simultaneously by an additive fractional Brownian motion (fBm) with Hurst parameter H \in (1/2,1) and a Poisson random measure, more realistic in modelling real world phenomena. Approximations in space are performed by the standard finite element method and in time by the stochastic Magnus-type integrator or the linear semi-implicit Euler method. We investigate the mean-square errors estimates of our fully discrete schemes and the results show how the convergence orders depend on the regularity of the initial data and the driven processes. To the best of our knowledge, these two schemes are the first numerical methods to approximate the non-autonomous semilinear stochastic partial differential equation (SPDE) driven simultaneously by an additive fractional Brownian motion with Hurst parameter H and a Poisson random measure.

Numerical Analysis

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to provide strong convergence results for a class of numerical approximations of non - autonomous semilinear stochastic partial differential equations (SPDEs) driven by additive fractional Brownian motion (fBm) and Poisson random measures. Specifically, the paper focuses on a class of non - autonomous SPDEs in the following form:

\[
\begin{cases}
dX(t)+A(t)X(t)dt = F(X(t))dt+\varphi(t)dB_H(t)+\int_{\chi}\psi(z)\tilde{N}(dz,dt), & t\in[0,T]\\
X(0)=X_0
\end{cases}
\]

where:
- \(X(t)\) is a stochastic process in the Hilbert space \(H = L^2(\Lambda,\mathbb{R})\).
- \(A(t)\) is a second - order linear operator, which may vary with time and generates an analytic semigroup \(S_s(t)=e^{-tA(s)}\).
- \(F\) is a nonlinear term.
- \(B_H(t)\) is a \(Q\)-cylindrical fractional Brownian motion with Hurst parameter \(H\in(\frac{1}{2},1)\).
- \(\tilde{N}(dz,dt)\) is a compensated Poisson random measure.
- \(X_0\) is the initial data and is a random variable.

The main contribution of the paper lies in proposing two numerical methods to approximate the solutions of the above - mentioned SPDEs and analyzing the strong convergence of these methods. These two methods are:
1. **Stochastic Magnus - type Integrator (SMTI)**
2. **Linear Semi - implicit Euler Method**

The authors studied the mean - square error estimates of these fully discrete schemes and showed how the convergence order depends on the regularity of the initial values and the driving processes. This is the first time that numerical methods capable of simultaneously handling non - autonomous SPDEs driven by additive fractional Brownian motion and Poisson random measures have been proposed.

Through these studies, the paper fills the gap in the existing literature regarding numerical methods for such non - autonomous SPDEs and provides a theoretical basis for practical applications.