Exploring Field Programmable Gate Array Architectures for Pascal's Simplex-based multinomial option pricing

In financial engineering, the speed and accuracy of option pricing models are critical for decision-making in fast-paced markets. This thesis explores the application of Field Programmable Gate Arrays (FPGAs) to accelerate the computation of option pricing models, specifically focusing on Pascal's Simplex method. Pascal's Simplex is chosen for its scalability in handling multi-asset options, which becomes increasingly vital as the complexity of financial instruments grows. FPGAs are selected due to their ability to execute parallel computations efficiently, offering significant speedups compared to traditional CPU-based systems while maintaining computational accuracy. This work introduces a novel FPGA architecture for Pascal's Simplex-based multinomial option pricing and presents a detailed comparative analysis of FPGA and CPU implementations. The results show that FPGAs provide substantial performance improvements in terms of speed and scalability, both for European and American style options, especially for pricing models with a high number of underlying assets. This research demonstrates that FPGAs can offer up to 43 times speedup over traditional methods while preserving accuracy, making them a powerful tool for real-time financial analytics. Furthermore, the architecture supports generalization for a varying number of assets underlying each option, and implementations on state of the art hardware are presented for up to 7 assets. By introducing a novel FPGA architecture for Pascal's Simplex-based multinomial option pricing and demonstrating up to 43 times speedup over traditional CPU methods while preserving accuracy, this thesis significantly contributes to advancing real-time financial analytics for complex multi-asset options.

Grantor dc:publisher

University College Cork

Year dc:date.issued

2025

Author dc:creator

• O'Mahony, Aidan T.

Advisors dc:contributor.advisor

• Popovici, Emanuel
• Hanzon, Bernard

• Fintech accelerators
• Field-Programmable Gate Arrays (FPGAs)
• Multi asset pricing
• Digital systems

Statement dc:rights

• © 2025, Aidan Timothy O'Mahony.

Licence dc:rights.uri

• https://creativecommons.org/licenses/by/4.0/

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/10468/17422

OAI identifier oai:identifier

oai:cora.ucc.ie:10468/17422