Weak convergence of a fully discrete approximation of a linear stochastic evolution equation with a positive-type memory term
In this paper we are interested in the numerical approximation of the marginal distributions of the Hilbert space valued solution of a stochastic Volterra equation driven by an additive Gaussian noise. This equation can be written in the abstract Itô form as $$ \dd X(t) + \left (\int_0^t b(t-s) A X(s) \, \dd s \right) \, \dd t = \dd W^{_Q}(t), t\in (0,T]; ~ X(0) =X_0\in H, $$ \noindent where WQ is a Q-Wiener process on the Hilbert space H and where the time kernel b is the locally integrable potential tρ−2, ρ∈(1,2), or slightly more general. The operator A is unbounded, linear, self-adjoint, and positive on H. Our main assumption concerning the noise term is that A(ν−1/ρ)/2Q1/2 is a Hilbert-Schmidt operator on H for some ν∈[0,1/ρ]. The numerical approximation is achieved via a standard continuous finite element method in space (parameter h) and an implicit Euler scheme and a Laplace convolution quadrature in time (parameter Δt=T/N). %Let XhN be the discrete solution at time T. Eventually let φ:H→R is such that D2φ is bounded on H but not necessarily bounded and suppose in addition that either its first derivative is bounded on H and X0∈L1(Ω) or φ=∥⋅∥2 and X0∈L2(Ω). We show that for φ:H→R twice continuously differentiable test function with bounded second derivative, $$ | \E φ(X^N_h) - \E φ(X(T)) | \leq C \ln \left(\frac{T}{h^{2/ρ} + Δt} \right) (Δt^{ρν} + h^{2ν}), $$ \noindent for any 0≤ν≤1/ρ. This is essentially twice the rate of strong convergence under the same regularity assumption on the noise.
