Abstract
The last decade has seen a significant growth in the financial industry. The recent widespread use of Internet technology has increased the accessibility of the general population to financial data, thereby increasing the average portfolio size. This increase, compounded by the need for accurate real-time results, has led to a rising demand for faster risk simulations. Often, accurately pricing widespread instruments, such as Collateralized Debt Obligations (CDOs), can take excessively long due to their multifactor assets dependency. We present a hardware implementation for a MultiFactor Gaussian Copula (MFGC) CDO pricing algorithm. Through a detailed benchmark exploration we demonstrate how reconfigurable hardware could be used to exploit fine-grain parallelism. Our results show that our implementation mapped onto a Xilinx Virtex 5 (XC5VSX50T) FPGA is over 71 times faster than corresponding software running on a single core 3.4 GHz Intel Xeon processor.
- Bank of International Settlements (BIS). 2008. Amounts outstanding of over-the-counter (OTC) derivatives. http://www.bis.org/statistics/otcder/dt1920a.pdf.Google Scholar
- Baxter, R., Booth, S., et al. 2007. Maxwell - a 64 FPGA supercomputer. In Proceedings of the 2nd NASA/ESA Conference on Adaptive Hardware and Systems (AHS’07). 287--294. Google Scholar
Digital Library
- CDX. 2008. Markit CDX indecies. http://www.markit.com.Google Scholar
- Dorfman, D. and Canning, D. 2007. The actuary’s high-performance computing challenge. In Windows in Financial Services.Google Scholar
- Glasserman, P. 2004. Monte Carlo Methods in Financial Engineering. Springer.Google Scholar
- Glasserman, P. and Suchintabandid, S. 2007. Correlation expansions for CDO pricing. J. Bank. Finan. 5, 1375--1398.Google Scholar
Cross Ref
- Goodman, L. and Fabozzi, F. 2002. Collateralized Debt Obligations. John Wiley & Sons.Google Scholar
- Hall, J. and White, A. 2004. Valuation of a CDO and an nth to default CDS without Monte Carlo simulation. J. Deriva. 12, 2, 8--23.Google Scholar
Cross Ref
- Haugh, M. and Loe, A. 2001. Computational challenges in portfolio management. Comput. Sci. Engin. 3, 8, 54--59. Google Scholar
Digital Library
- Irturk, A., Benson, B., Laptev, N., and Kastner, R. 2008. FPGA acceleration of mean variance framework for optimal asset allocation. In Proceedings of the Workshoop on High Performance Computational Finance (WHPCF’08). 1--8.Google Scholar
- Jackson, K., Kreinin, A., and Ma, X. 2007. Loss distribution evaluation for synthetic CDOs. http://www.defaultrisk.com/pp_cdo_14.htm.Google Scholar
- Kaganov, A., Lakhany, A., and Chow, P. 2008. FPGA acceleration of Monte Carlo based credit derivative pricing. In Proceedings of the International Conference on Field Programmable Logic and Applications. 329--334.Google Scholar
- Lee, D., Luk, W., Villasenor, J., Zhang, G., and Leong, P. 2005. A hardware Gaussian noise generator using the Wallace method. IEEE Trans. VLSE Syst. 13, 8, 911--920. Google Scholar
Digital Library
- Li, D. 2000. On default correlation: A copula function approach. J. Fixed Income 9, 43--54.Google Scholar
Cross Ref
- McCool, M., Wadleigh, K., Henderson, B., and Lin, H. 2006. Performance evaluation of GPUs using rapidmind development platform. In Proceedings of the ACM/IEEE Conference on Supercomputing. Google Scholar
Digital Library
- Morris, G. and Aubury, M. 2007. Design space exploration of the European option benchmark using hyperstreams. In Proceedings of the Field Programmable Logic and Applications Conference (FPL’07). 5--10.Google Scholar
- Przybus, B. 2008. The virtex-5 SXT option for high-performance digital signal processing. http://www.xilinx.com/publications/xcellonline/xcell_62/xc_pdf/p34-36_62-virtex.pdf.Google Scholar
- SIFMA. 2008. Global market issuance data. http:/www.sifma.org.Google Scholar
- Thomas, D., Bower, J., and Luk, W. 2007. Automatic generation and optimization of reconfigurable financial Monte-Carlo simulations. In Proceedings of the Conference Application-Specific Systems, Architectures and Processors. 168--173.Google Scholar
- Thomas, D. and Luk, W. 2007. Sampling from the multivariate Gaussian distribution using reconfigurable hardware. In Proceedings of the Conference Field-Programmable Custom Computing Machines (FCCM07). 3--12. Google Scholar
Digital Library
- Xilinx, Inc. 2009. Fast simplex link (FSL) bus (v2.11b). http://www.xilinx.com/support/documentation/ip_documentation/fsl_v20.pdf.Google Scholar
- Zhang, G., Leong, P., Ho, C., Tsoi, K., Cheung, C., Lee, D., Cheung, R., and Luk, W. 2005. Reconfigurable acceleration for Monte Carlo based financial simulation. In Proceedings of the Conference Field-Programmable Technology. 215--222.Google Scholar
Index Terms
FPGA Acceleration of MultiFactor CDO Pricing
Recommendations
FPGA acceleration of CDO pricing based on correlation expansions
ACM SIGARCH Computer Architecture News/HEART '12Because of the significant growth in the financial market, faster and accurate pricing of widespread instruments is becoming more important. In this paper, we describe an FPGA implementation of an analytical method for collateralized debt obligation (...
On Comparing Financial Option Price Solvers on FPGA
FCCM '11: Proceedings of the 2011 IEEE 19th Annual International Symposium on Field-Programmable Custom Computing MachinesA number of different numerical methods for accelerating financial option pricing using FPGAs have recently been investigated, such as Monte-Carlo, finite-difference, quadrature, and binomial trees. However, these papers only compare acceleration of ...
Unifying Finite Difference Option-Pricing for Hardware Acceleration
FPL '11: Proceedings of the 2011 21st International Conference on Field Programmable Logic and ApplicationsExplicit finite difference method is widely used in finance for pricing many kinds of options. Its regular computational pattern makes it an ideal candidate for acceleration using reconfigurable hardware. However, because the corresponding hardware ...






Comments