Monte Carlo versus multilevel Monte Carlo in weak error simulations of SPDE approximations

Journal article, 2018

The simulation of the expectation of a stochastic quantity E[Y] by Monte Carlo methods is known to be computationally expensive especially if the stochastic quantity or its approximation Y-n is expensive to simulate, e.g., the solution of a stochastic partial differential equation. If the convergence of Y-n to Y in terms of the error |E[Y - Y-n]| is to be simulated, this will typically be done by a Monte Carlo method, i.e., |E[Y] - E-N [Y-n]| is computed. In this article upper and lower bounds for the additional error caused by this are determined and compared to those of |E-N [Y - Y-n]|, which are found to be smaller. Furthermore, the corresponding results for multilevel Monte Carlo estimators, for which the additional sampling error converges with the same rate as |E[Y - Y-n]|, are presented. Simulations of a stochastic heat equation driven by multiplicative Wiener noise and a geometric Brownian motion are performed which confirm the theoretical results and show the consequences of the presented theory for weak error simulations.