Multilevel Monte Carlo for Lévy-driven SDEs: central limit theorems for adaptive Euler schemes (Q259571)

The authors extend the multilevel Monte Carlo method for the numerical computation of expectations for stochastic differential equations (SDEs), introduced in [\textit{M. B. Giles} Oper. Res. 56, No. 3, 607--617 (2008; Zbl 1167.65316)], to stochastic differential equations driven by Lévy processes. Based on a jump-adapted Euler scheme, various implementable schemes are discussed. As a main result, they prove a stable convergence for the normalized error process. Further main results are central limit theorems for path-dependent functionals, including functionals on integral averages and supremum-dependent functionals. These central limit theorems are utilized to optimise the parameters in the multilevel scheme.