Abstract
The Heston model is the most renowned stochastic volatility function in finance, but the calibration input parameters is a challenging task. This contest grows up because the instantaneous volatility is unobservable or market quotes are absent/unviable to agents, remaining the asset time-series as tangible. Moreover, these conditions are unfit to approach based on Maximum Likelihood Estimation or optimisation of the differences in Pricing models and quotes in the real market. Today, neural networks are a well-known tool with malleable and powerful features to map accurately any nonlinear continuous operator in complex systems. This work adopts the deep operator networks (DeepONets) to learn the Heston parameters based on observed time series. We perform simulations of trajectories following the Heston model with a truncated Euler discretization scheme and randomised inputs parameters. The five parameters are estimated by Stacked and Unstacked DeepONets and compared with GJR-GARCH and standard neural network with the Tukeys’ test. The results indicated the improvement in accuracy of the Unstacked model. However, the statistical test indicates some similarity between the Unstacked model and standard neural network.
Supported by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior-CAPES, Fundação de Amparo à Pesquisa do Estado de Minas Gerais-FAPEMIG (grant TEC-APQ-00334/18) and Conselho Nacional de Desenvolvimento Científico e Tecnológico-CNPq (grant 429639/2016-3).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The problem of parameter estimation for stochastic processes is not limited to the Heston model. This gap is shown in [21] with exemplification for interest rates.
- 2.
See [7, Chapter 6.4.1].
- 3.
See [11][Chapter 15, Sect. 11] for more details about implied volatilities.
- 4.
The GJR-GARCH model also captures the stylised facts in finance, like the GARCH model, but appends the relation between negative shocks returns at last observation as positive shocks. This asymmetry is known as the leverage effect.
- 5.
CIR is a short-form of Cox, Ingersoll, and Ross (the original authors), see [6, Chapter 2, pg. 15].
- 6.
Mrázek and Pospíšil [23] describes some discretization and truncation schemes with richness of details.
- 7.
These interval was choice to comprise the values observed in literature. However, the plenty combinations of parameters are feasible depending of the market considered.
- 8.
The price time-series is convert in return time-series, it is a common procedure in finance and can be interpreted as the standardisation step.
- 9.
Linear activation was considered in output node to \(\alpha \) and \(\rho \). The values inside brackets indicate the number of neurons in each layer. Other topologies were tested, but without significant gain.
- 10.
A Python library designed for scientific machine learning (https://github.com/lululxvi/deepxde).
References
Barandas, M., et al: Tsfel: Time series feature extraction library. SoftwareX 11, 100456 (2020). https://doi.org/10.1016/j.softx.2020.100456
Cai, S., Wang, Z., Lu, L., Zaki, T.A., Karniadakis, G.E.: Deepm&mnet: inferring the electroconvection multiphysics fields based on operator approximation by neural networks. J. Comput. Phys. 436, 110296 (2021). https://doi.org/10.1016/j.jcp.2021.110296
Cape, J., Dearden, W., Gamber, W., Liebner, J., Lu, Q., Nguyen, M.L.: Estimating heston’s and bates’ models parameters using Markov chain monte carlo simulation. J. Stat. Comput. Simul. 85(11), 2295–2314 (2015). https://doi.org/10.1080/00949655.2014.926899
Engle, R.F., Lee, G.G.J.: Estimating diffusion models of stochastic volatility. In: Rossi, P. (ed.) MODELLING STOCK MARKET VOLATILITY: Bridging the Gap to Continuous Time, vol. 1, chap. 11, pp. 333–355. Academic Press Inc, 525 B Street, Suite 1900, San Diego, California 92101–4495, USA (1996)
Gallant, A.R., Tauchen, G.: Which moments to match? Econometric Theor. 12(4), 657–681 (1996). https://doi.org/10.1017/S0266466600006976
Gatheral, J.: The volatility surface : a practitioner’s guide. Wiley (2012). https://doi.org/10.1002/9781119202073
Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. Adaptive Computation and Machine Learning Series. MIT Press, Cambridge (2017)
Gourieroux, C., Monfort, A., Renault, E.: Indirect inference. J. Appl. Econ. 8, S85–S118 (1993). http://www.jstor.org/stable/2285076
Hernandez, A.: Model calibration with neural networks (2015). https://doi.org/10.2139/ssrn.2812140
Heston, S.L.: A closed-form solution for options with stochastic volatility with applications to bond and currency options. Rev. Finan. Stud. 6(2), 327–343 (2015). https://doi.org/10.1093/rfs/6.2.327
Hull, J.: Options, Futures, and Other Derivatives. 10th edn, Pearson Prentice Hall, Upper Saddle River (2017)
Jacquier, E., Polson, N.G., Rossi, P.E.: Bayesian analysis of stochastic volatility models. J. Bus. Econ. Stat. 12(4), 371–389 (1994). http://www.jstor.org/stable/1392199
Kloeden, P.E., Platen, E.P.: Stochastic Modelling and Applied Probability, Applications of Mathematics, 2 edn. vol. 1. Springer, Berlin (1992)
Lewis, A.L.: Option Valuation Under Stochastic Volatility: With Mathematica Code, chap. Appendix 1.1 - Parameter Estimators for the GARCH Diffusion Model. Finance Press, Newport Beach, California, USA (2000)
Liu, S., Borovykh, A., Grzelak, L.A., Oosterlee, C.W.: A neural network-based framework for financial model calibration. J. Math. Ind. 9(1), 9 (2019). https://doi.org/10.1186/s13362-019-0066-7
Lu, L., Jin, P., Karniadakis, G.E.: Deeponet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators (2020). https://arxiv.org/abs/1910.03193
Lu, L., Jin, P., Pang, G., Zhang, Z., Karniadakis, G.E.: Learning nonlinear operators via deeponet based on the universal approximation theorem of operators. Nature Mach. Intell. 3(3), 218–229 (2021). https://doi.org/10.1038/s42256-021-00302-5
Lu, L., Meng, X., Mao, Z., Karniadakis, G.E.: DeepXDE: a deep learning library for solving differential equations. SIAM Rev. 63(1), 208–228 (2021). https://doi.org/10.1137/19M1274067
Mao, Z., Lu, L., Marxen, O., Zaki, T.A., Karniadakis, G.E.: Deepm & mnet for hypersonics: Predicting the coupled flow and finite-rate chemistry behind a normal shock using neural-network approximation of operators (2020). https://arxiv.org/abs/2011.03349
Márkus, L., Kumar, A.: Modelling joint behaviour of asset prices using stochastic correlation. Method. Comput. Appl. Probab. 23(1), 341–354 (2021). https://doi.org/10.1007/s11009-020-09838-2
Monsalve-Cobis, A., González-Manteiga, W., Febrero-Bande, M.: Goodness-of-fit test for interest rate models: an approach based on empirical processes. Comput. Stat. Data Anal. 55(12), 3073–3092 (2011). https://doi.org/10.1016/j.csda.2011.06.004
Moysiadis, G., Anagnostou, I., Kandhai, D.: Calibrating the mean-reversion parameter in the hull-white model using neural networks. In: Alzate, C., et al. (eds.) MIDAS/PAP-2018. LNCS (LNAI), vol. 11054, pp. 23–36. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-13463-1_2
Mrázek, M., Pospíšil, J.: Calibration and simulation of heston model. Open Math. 15(1), 679–704 (2017). https://doi.org/10.1515/math-2017-0058
Rouah, F.D.: The Heston model and its extensions in Matlab and C#, Wiley finance series, 1 edn, vol. 1. Wiley, Hoboken (2013)
Tang, C.Y., Chen, S.X.: Parameter estimation and bias correction for diffusion processes. J. Econ. 149(1), 65–81 (2009). https://doi.org/10.1016/j.jeconom.2008.11.001
Tomas, M.: Pricing and calibration of stochastic models via neural networks. Master’s thesis, Department of Mathematics, Imperial College London (2018). https://www.imperial.ac.uk/media/imperial-college/faculty-of-natural-sciences/department-of-mathematics/math-finance/TOMAS_MEHDI_01390785.pdf
Wang, X., He, X., Bao, Y., Zhao, Y.: Parameter estimates of heston stochastic volatility model with mle and consistent ekf algorithm. Sci. China Inf. Sci. 61(4), 042202 (2018). https://doi.org/10.1007/s11432-017-9215-8
Xie, Z., Kulasiri, D., Samarasinghe, S., Rajanayaka, C.: The estimation of parameters for stochastic differential equations using neural networks. Inverse Prob. Sci. Eng. 15(6), 629–641 (2007). https://doi.org/10.1080/17415970600907429
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Leite, I.M.S., Yamim, J.D.M., da Fonseca, L.G. (2021). The DeepONets for Finance: An Approach to Calibrate the Heston Model. In: Marreiros, G., Melo, F.S., Lau, N., Lopes Cardoso, H., Reis, L.P. (eds) Progress in Artificial Intelligence. EPIA 2021. Lecture Notes in Computer Science(), vol 12981. Springer, Cham. https://doi.org/10.1007/978-3-030-86230-5_28
Download citation
DOI: https://doi.org/10.1007/978-3-030-86230-5_28
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-86229-9
Online ISBN: 978-3-030-86230-5
eBook Packages: Computer ScienceComputer Science (R0)