PARALLEL COMPUTING METHOD OF VALUING FOR MULTI-ASSET EUROPEAN OPTION

Abstract

A critical problem in finance engineering is to value the option and other derivatives securities correctly. The Monte Carlo method (MC) is an important one in the computation for the valuation of multi-asset European option. But its convergence rate is very slow. So various quasi Monte Carlo methods and the relative parallel computing method are becoming an important approach to the valuing of multi-asset European option. In this paper, we use a number-theoretic method, which is a H–W method, to generate identical distributed point set in order to compute the value of the multi-asset European option. It turns out to be very effective, and the time of computing is greatly shortened. Comparing with other methods, the method computes less points and it is especially suitable for high dimension problem.