A multi-point distributed random variable accelerator for Monte Carlo simulation in finance

Abstract

The pricing and hedging of complex derivative securities via Monte Carlo simulations of stochastic differential equations constitutes an intensive computational task. To achieve "real time" execution, as often required by financial institutions, one needs highly efficient implementations of the multi-point distributed random variables underlying the simulations. In this paper a fast and flexible dedicated hardware solution is proposed. A comparative performance analysis demonstrates that the hardware solution is bottleneck-free and flexible, and significantly increases the computational efficiency of the software solution.