Convergence of the Modified Craig-Sneyd scheme for two-dimensional convection-diffusion equations with mixed derivative term
This provides a theoretical foundation for a widely used numerical method in financial mathematics and other fields, but the result is incremental as it extends existing analysis to a specific scheme.
The paper proves the first convergence theorem for the Modified Craig-Sneyd scheme applied to two-dimensional convection-diffusion equations with mixed derivative terms, establishing a global temporal error bound independent of spatial mesh width. Numerical experiments confirm the result.
We consider the Modified Craig-Sneyd (MCS) scheme which forms a prominent time stepping method of the Alternating Direction Implicit type for multidimensional time-dependent convection-diffusion equations with mixed spatial derivative terms. Such equations arise often, notably, in the field of financial mathematics. In this paper a first convergence theorem for the MCS scheme is proved where the obtained bound on the global temporal discretization errors has the essential property that it is independent of the (arbitrarily small) spatial mesh width from the semidiscretization. The obtained theorem is directly pertinent to two-dimensional convection-diffusion equations with mixed derivative term. Numerical experiments are provided that illustrate our result.