Hardware Acceleration of Monte - Carlo Structural Financial Instrument Pricing Using a Gaussian Copula Model

In recent years the financial world has seen an increasing demand for faster risk simulations, driven by increasing contract complexity and client portfolio growth. Traditionally many financial models employ Monte-Carlo simulation, which can take excessively long to compute in software. Hence, commonly a hardware accelerator is sought out. This thesis focuses on accelerating structured financial instruments, namely Collateralized Debt Obligations (CDOs) pricing, which have previously not been targeted for hardware acceleration despite their prominence in the financial market. This thesis presents a hardware implementation for the One-Factor and the Multi-Factor Gaussian Copula models. It also explores the precision requirements and the resulting resource utilization for each such numerical representation. The results show that the hardware implementation mapped onto a Xilinx XC5VSX50T chip is over 64 and 71 times faster than corresponding software running on a 3.4 GHz Intel Xeon processor, for the One-Factor and the Multi-Factor models, respectively.