ABSTRACT

Computationally expensive time-stepping is the bottleneck of finite-difference methods used for valuing multi-asset options. The authors consider a novel algorithm with radically accelerated convergence, which is based on an optimal approximation of the matrix exponential. This algorithm is modified to compute the price of American options. A reduction of one order in computational time compared with one of the most efficient variants of the Crank-Nicolson scheme is observed with a two-factor convertible bond model.