×

XVA principles, nested Monte Carlo strategies, and GPU optimizations. (English) Zbl 1416.91398

Summary: We present a nested Monte Carlo (NMC) approach implemented on graphics processing units (GPUs) to X-valuation adjustments (XVAs), where X ranges over C for credit, F for funding, M for margin, and K for capital. The overall XVA suite involves five compound layers of dependence. Higher layers are launched first, and trigger nested simulations on-the-fly whenever required in order to compute an item from a lower layer. If the user is only interested in some of the XVA components, then only the sub-tree corresponding to the most outer XVA needs be processed computationally. Inner layers only need a square root number of simulation with respect to the most outer layer. Some of the layers exhibit a smaller variance. As a result, with GPUs at least, error-controlled NMC XVA computations are doable. But, although NMC is naively suited to parallelization, a GPU implementation of NMC XVA computations requires various optimizations. This is illustrated on XVA computations involving equities, interest rate, and credit derivatives, for both bilateral and central clearing XVA metrics.

MSC:

91G60 Numerical methods (including Monte Carlo methods)
65C05 Monte Carlo methods
91-04 Software, source code, etc. for problems pertaining to game theory, economics, and finance
91-08 Computational methods for problems pertaining to game theory, economics, and finance

Software:

CUDA
Full Text: DOI

References:

[1] Abbas-Turki, L. A.; Bouselmi, A. I.; Mikou, M. A., Toward a coherent Monte Carlo simulation of CVA, Monte Carlo Methods Applications, 20, 3, 195-216, (2014) · Zbl 1302.91191
[2] Abbas-Turki, L. A.; Graillat, S., Resolving small random symmetric linear systems on graphics processing units, Journal of Supercomputing, 73, 4, 1360-1386, (2017)
[3] L. Abbas-Turki & M. Mikou (2015) TVA on American derivatives, hal-01142874 (Working Paper).
[4] Abbas-Turki, L.; Vialle, S.; Lapreye, B.; Mercier, P., High dimensional pricing of exotic European contracts on a GPU cluster, and comparison to a CPU cluster, IEEE IPDPS: Parallel and Distributed Computing in Finance, (2009)
[5] Abbas-Turki, L. A.; Vialle, S.; Lapeyre, B.; Mercier, P., Pricing derivatives on graphics processing units using Monte Carlo simulation, Concurrency and Comput. Pract. Exp., 26, 9, 1679-1697, (2014)
[6] C. Albanese & L. Andersen (2014) Accounting for OTC derivatives: Funding adjustments and the re-hypothecation option, https://papers.ssrn.com/sol3/papers.cfm?abstract˙id=2482955.
[7] C. Albanese & L. Andersen (2015) FVA: What’s wrong, and how to fix it. Risk Magazine 54-56.
[8] C. Albanese, L. Andersen & S. Iabichino (2015) FVA: Accounting and risk management, Risk Magazine 64-68.
[9] Albanese, C.; Bellaj, T.; Gimonet, G.; Pietronero, G., Coherent global market simulations and securitization measures for counterparty credit risk, Quantitative Finance, 11, 1, 1-20, (2011) · Zbl 1217.91194
[10] Albanese, C.; Caenazzo, S.; Crépey, S., Credit, funding, margin, and capital valuation adjustments for bilateral portfolios, Probability, Uncertainty and Quantitative Risk, 2, 7, 26, (2017) · Zbl 1432.91130
[11] C. Albanese & S. Crépey (2017) XVA analysis from the balance sheet, https://math.maths.univ-evry.fr/crepey.
[12] C. Albanese & H. Li (2010) Monte Carlo pricing using operator methods and measure changes, https://papers.ssrn.com/sol3/papers.cfm?abstract˙id=1484556.
[13] An, J.; Owen, A., Quasi-regression, Journal of Complexity, 17, 4, 588-607, (2001) · Zbl 0993.65018
[14] L. Andersen, D. Duffie & Y. Song (2017) Funding value adjustments, Journal of Finance, Forthcoming.
[15] F. Anfuso, D. Aziz, K. Loukopoulos & P. Giltinan (2017) A sound modeling and backtesting framework for forecasting initial margin requirements, Risk Magazine 1-6.
[16] Y. Armenti & S. Crépey (2017) XVA Metrics for CCP optimizsation, https://math.maths.univ-evry.fr/crepey.
[17] Bichuch, M.; Capponi, A.; Sturm, S., Arbitrage-free XVA, Mathematical Finance, 28, 2, 582-620, (2018) · Zbl 1390.91276
[18] Brigo, D.; Pallavicini, A., Nonlinear consistent valuation of CCP cleared or CSA bilateral trades with initial margins under credit, funding and wrong-way risks, Journal of Financial Engineering, 1, 1-60, (2014)
[19] Cesari, G.; Aquilina, J.; Charpillon, N.; Filipovic, Z.; Lee, G.; Manda, I., Modeling, Pricing, and Hedging Counterparty Credit Exposure, (2010), Springer Finance
[20] Clément, E.; Lamberton, D.; Protter, P., An analysis of a least squares regression algorithm for American option pricing, Finance and Stochastics, 17, 448-471, (2002) · Zbl 1039.91020
[21] Coates, A.; Huval, B.; Wang, T.; Wu, D.; Catanzaro, B.; Andrew, N., Deep learning with COTS HPC systems, Proceedings of the 30th International Conference on Machine Learning, 28, 1337-1345, (2013)
[22] Crépey, S., Financial Modeling: A Backward Stochastic Differential Equations Perspective, (2013), Springer Finance Textbooks · Zbl 1271.91004
[23] Crépey, S.; Bielecki, T. R.; Brigo, D., Counterparty Risk and Funding: A Tale of Two Puzzles, Chapman & Hall/CRC Financial Mathematics Series · Zbl 1294.91005
[24] S. Crépey, R. Élie, W. Sabbagh & S. Song (2017) When capital is a funding source: The XVA Anticipated BSDEs, https://math.maths.univ-evry.fr/crepey.
[25] S. Crépey, G. Fort, E. Gobet & U. Stazhynski (2017) Uncertainty quantification for stochastic approximation limits using chaos expansion, hal-01629952.
[26] Crépey, S.; Nguyen, T. M., Challenges in Derivatives Markets, Nonlinear Monte carl schemes for counterparty risk on credit derivatives, 53-82, (2016), Springer · Zbl 1398.91668
[27] Crépey, S.; Rahal, A., Simulation/regression pricing schemes for CVA computations on CDO tranches, Communications in Statistics-Theory and Methods, 43, 7, 1390-1408, (2014) · Zbl 1290.91178
[28] Crépey, S.; Song, S., Counterparty risk and funding: immersion and beyond, Finance and Stochastics, 20, 4, 901-930, (2016) · Zbl 1380.91139
[29] Delmas, J.-F.; Jourdain, B., Mathématiques et Applications, 57, Modèles aléatoires, (2006), Springer
[30] Elouerkhaoui, Y., Pricing and hedging in a dynamic credit model, International Journal of Theoretical and Applied Finance, 10, 4, 703-731, (2007) · Zbl 1291.91223
[31] Y. Elouerkhaoui (2016) From FVA to KVA: Including cost of capital in derivatives pricing, Risk Magazine 70-75.
[32] Elouerkhaoui, Y., Credit Correlation: Theory and Practice, (2017), Palgrave Macmillan
[33] Embrechts, P.; Klueppelberg, C.; Mikosch, T., Applications of Mathematics, 33, Modeling extremal events for insurance and finance, (1997), Springer
[34] Fujii, M.; Takahashi, A., Analytical approximation for nonlinear FBSDEs with perturbation scheme, International Journal of Theoretical and Applied Finance, 15, 5, 1250034, (2012) · Zbl 1262.91159
[35] Fujii, M.; Takahashi, A., Perturbative expansion of FBSDE in an incomplete market with stochastic volatility, Quarterly Journal of Finance, 2, 3, 1250015, (2012)
[36] Gerhold, S., The longstaff-Schwartz algorithm for Lévy models: results on fast and slow convergence, The Annals of Applied Probability, 21, 2, 589-608, (2011) · Zbl 1219.62161
[37] Giles, M., Multilevel Monte Carlo path simulation, Operations Research, 56, 607-617, (2008) · Zbl 1167.65316
[38] Gobet, E.; Lemor, J.-P.; Warin, X., A regression-based Monte Carlo method to solve backward stochastic differential equations, The Annals of Applied Probability, 15, 3, 2172-2202, (2005) · Zbl 1083.60047
[39] Gobet, E.; Pagliarani, S., Analytical approximations of BSDEs with nonsmooth driver, SIAM Journal on Financial Mathematics, 6, 919-958, (2015) · Zbl 1326.65019
[40] Gordy, M. B.; Juneja, S., Nested simulation in portfolio risk measurement, Management Science, 56, 10, 1833-1848, (2010) · Zbl 1232.91622
[41] Green, A., XVA: Credit, Funding and Capital Valuation Adjustments., (2015), Wiley
[42] A. Green & C. Kenyon (2015) MVA by replication and regression, Risk Magazine 82-87.
[43] A. Green, C. Kenyon & C. Dennis (2014) KVA: Capital valuation adjustment by replication, Risk Magazine, 82-87, ssrn.2400324.
[44] Green, O.; McColl, R.; Bader, D. A., GPU merge path — A GPU merging algorithm, 26th ACM International Conference on Supercomputing (ICS), 25-29, (2012), The ACM Digital Library
[45] P. Henry-Labordère (2012) Cutting CVA’s complexity, Risk Magazine 67-73.
[46] Hull, J.; White, A., CVA and wrong way risk, Financial Analyst Journal, 68, 58-69, (2012)
[47] L’Ecuyer, P., Combined multiple recursive random number generators, Operations Research, 44, 5, (1996) · Zbl 0904.65004
[48] M. Li & F. Mercurio (2015) Jumping with default: Wrong-way risk modeling for CVA, Risk Magazine.
[49] Li, S.; Amenta, N., Similarity Search and Applications, Brute-force k-nearest neighbors search on the GPU, 259-270, (2015), Springer
[50] Longstaff, F. A.; Schwartz, E. S., Valuing American options by simulation: A simple least-squares approach, The Review of Financial Studies, 14, 1, 113-147, (2001)
[51] Moran, L.; Wilkens, S., Capturing initial margin in counterparty risk calculations, Journal of Risk Management in Financial Institutions, 10, 118-129, (2017)
[52] Newey, W. K., Convergence rates and asymptotic normality for series estimators, Journal of Econometrics, 79, 147-168, (1995) · Zbl 0873.62049
[53] Nvidia (2017a) Cuda C best practices guide.
[54] Nvidia (2017b) Cuda C programming guide.
[55] Nvidia (2017c) Nvidia NVLink TM high-speed interconnect: Application performance.
[56] Pykhtin, M., General wrong-way risk and stress calibration of exposure, Journal of Risk Management in Financial Institution, 5, 234-251, (2012)
[57] T. Rainforth, R. Cornish, H. Yang, A. Warrington & F. Wood (2017) On the opportunities and pitfalls of nesting monte carlo estimators, arXiv:1709.06181v3.
[58] A. Reghai, O. Kettani & M. Messaoud (2015) CVA with Greeks and AAD, Risk Magazine.
[59] Singh, D. P.; Joshi, I.; Choudhary, J., Survey of GPU-based sorting algorithms, International Journal of Parallel Programming, (2017)
[60] M. I. Taarit (2017) Pricing of XVA Adjustments: From Expected Exposures to Wrong-Way risks, Ph.D Thesis, Université Paris-Est, Forthcoming.
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.