Moment matching machine learning methods for risk management of large variable annuity portfolios. (English) Zbl 1401.91525

Summary: Variable annuity (VA) with embedded guarantees have rapidly grown in popularity around the world in recent years. Valuation of VAs has been studied extensively in past decades. However, most of the studies focus on a single contract. These methods can be extended neither to valuate nor to manage the risk of a large variable annuity portfolio due to the computational complexity. In this paper, we propose an efficient moment matching machine learning method to solve this problem. This method is proved to be a good candidate for risk management in terms of the speed of and the complexity of computing the annual dollar deltas, VaRs and CVaRs for a large variable annuity portfolio whose contracts are over a period of 25 years. There are two stages for our method. First, we select a small number of contracts and propose a moment matching Monte Carlo method based on the Johnson curve, rather than the well known nested simulations, to compute the annual dollar deltas, VaRs and CVaRs for each selected contract. Then, these computed results are used as a training set for well known machine learning methods, such as regression tree, neural network and so on. Afterwards, the annual dollar deltas, VaRs and CVaRs for the entire portfolio are predicted through the trained machine learning method. Compared to other existing methods [D. Bauer et al., Astin Bull. 38, No. 2, 621–651 (2008; Zbl 1274.91399); G. Gan, Insur. Math. Econ. 53, No. 3, 795–801 (2013; Zbl 1290.91086); with X. S. Lin, ibid. 62, 138–150 (2015; Zbl 1318.91112)], our method is very efficient and accurate. Finally, our test results support our claims.


91G10 Portfolio theory
62P05 Applications of statistics to actuarial sciences and financial mathematics


Full Text: DOI


[1] Armstrong, M., The reset decision for segregated fund maturity guarantees, Insurance: Math. Econ., 29, 257-269, (2001) · Zbl 0990.91027
[2] Bauer, D.; Kling, A.; Russ, J., A universal pricing framework for guaranteed minimum benefits in vairable annuities, ASTIN Bull., 38, 621-651, (2008) · Zbl 1274.91399
[3] Bauer, D.; Russ, A.; Singer, D., On the calculation of the solvency capital requirement based on nested simulations, ASTIN Bull., 42, 453-499, (2012) · Zbl 1277.91074
[4] Belanger, A.; Forsyth, P.; Labahn, G., Valuing the guaranteed minimum death benefit clause with partial withdraws, Appl. Math. Finance, 16, 451-496, (2009) · Zbl 1189.91066
[5] Bishop, C. M., Neural networks for pattern recognition, (1995), Oxford University Press
[6] Boyle, P.; Kolkiewicz, A.; Tan, K., Valuation of the reset options embedded in some equity-link insurance products, North Am. Actuarial J., 5, 1-18, (2001) · Zbl 1083.91511
[7] Boyle, P.; Tian, W., The design of equity-indexed annuities, Insurance: Math. Econ., 43, 303-315, (2008) · Zbl 1152.91484
[8] Braspenning, P. J.; Thuijsman, F.; Weijters, A. J.M. M., Artificial neural networks: an introduction to ANN theory and practice, 931, (1995), Springer Science & Business Media · Zbl 1153.68302
[9] Breiman, L., Random forests, Mach. Learn., 45, 1, 5-32, (2001) · Zbl 1007.68152
[10] Cathcart, M.; Morrison, S., Variable annuity economic capital: the least squares Monte Carlo approach, Life Pension, 36-40, (2009)
[11] Coleman, T.; Li, Y.; Patron, M., Pricing American options on jump-diffusion processes using Fourier Hermite series expansions, Insurance: Math Econ., 38, 215-228, (2006)
[12] Coleman, T.; Xu, W., Automatic differentiation in MATLAB using ADMAT (with applications), (2016), SIAM, Pheladephia · Zbl 1364.65056
[13] Das, S.; Sundaram, R., Of smiles and smirks: a term structure perspective, J. Financ. Quantit. Anal., 34, 02, 211-239, (1999)
[14] Daul, S.; Vidal, E., Replication of insurance liabilities, RiskMetrics J., 9, (2009)
[15] Dembo, R.; Rosen, D., The practice of portfolio replication: a practical overview of forward and inverse problems, Ann. Oper. Res., 85, 267-284, (1999) · Zbl 0920.90006
[16] Duan, J.; Gauthier, G.; Sasseville, C.; Simonato, J., Approximating the gjr-garch and egarch option pricing models analytically, J. Comput. Finance, 9, 41-69, (2006)
[17] Duan, J.; Gauthier, G.; Simonato, J., An analytical approximation for the garch option pricing model, J. Comput. Finance, 2, 75-116, (1999)
[18] Fox, J., A nested approach to simulation var using mosec, Insights, 1-7, (2013)
[19] Gan, G., Application of data clustering and machine learning in variable annuity valuation, Insurance: Math. Econ., 53, 795-801, (2013) · Zbl 1290.91086
[20] Gan, G.; Lin, X., Valuation of large variable annuity portfolios under nested simulation: a functional data approach, Insurance: Math. Econ., 62, 138-150, (2015) · Zbl 1318.91112
[21] Gao, J.; Ulm, E., Optimal consumption and allocation in variable annuities with guaranteed minimum death benefits, Insurance: Math. Econ., 51, 3, 586-598, (2012) · Zbl 1285.91056
[22] Giles, M.; Glasserman, P., Computation methods: smoking adjoints: fast Monte Carlo greeks, Risk, 88-92, (2006)
[23] Hardy, M., A regime-switching model of long-tem stock returns, North Am. Actuarial J., 5, 41-53, (2001) · Zbl 1083.62530
[24] Heston, S.; Nandi, S., A closed-form garch option pricing model, Rev. Financ. Studies, 13, 586-625, (2000)
[25] Hill, I. D.; Hill, R.; Holder, R. L., Algorithm as 99: Fitting Johnson curves by moments, Appl. Stat., 180-189, (1976)
[26] Huang, H.; Milevsky, M.; Salisbury, T., Optimal initiation of a glwb in a variable annuity: no arbitrage approach, Insurance: Math. Econ., 56, 102-111, (2014) · Zbl 1304.91113
[27] Hull, J., Options, futures, and other derivatives, (2006), Pearson Education · Zbl 1087.91025
[28] Johnson, N., System of frequency curves generated by methods of translation, Biometrika, 36, 149-176, (1949) · Zbl 0033.07204
[29] Kou, S., A jump-diffusion model for option pricing, Manage Sci., 48, 1086-1101, (2002) · Zbl 1216.91039
[30] Lin, X.; Tan, K.; Yang, H., Pricing annuity guarantees under a regime switching model, North Am. Actuarial J., 13, 316-338, (2009)
[31] Longstaff, F.; Schwartz, E., Valuing American options by simulation: a simple least squares approach, Rev. Financ. Study, 14, 113-147, (2001)
[32] Lopes, N.; Ribeiro, B., Machine learning for adaptive many-core machines-A practical approach, (2015), Springer
[33] Mazzoni, T., Fast analytic option valuation with garch, J. Derivatives, 18, 18-40, (2010)
[34] Merton, R., Hedging guarantees in variable annuities under both equity and interest rate risk, J Financ. Econ., 3, 124-144, (1976)
[35] Milevsky, M.; Promislow, S., Mortality derivatives and the option to annuities, Insurance: Math. Econ., 9, 299-318, (2001) · Zbl 1074.62530
[36] Ng, A. C.-Y.; Li, J. S.-H., Pricing and hedging variable annuity guarantees with multiasset stochastic investment models, North Am. Actuarial J., 17, 41-62, (2013)
[37] Olshen, L.; Stone, C. J., Classification and regression trees, Wadsworth International Group, 93, 99, 101, (1984)
[38] Reynolds, C.; Man, S., Nested stochastic pricing: the time has come, Product Matters! - Soc. Actuaries, 71, 16-20, (2008)
[39] Sloane, W., Life insurers, variable annuities and mutual funds: a critical study, J. Risk Insurance, 37, 99, (1970)
[40] W. Xu; Chen, X.; Coleman, T., The efficient application of automatic differentiation for computing gradients in financial applications, J. Comput. Finance, 19, 3, 71-96, (2016)
[41] Xu, W.; Hong, Z.; Qin, C., A new sampling strategy willow tree method with application to path-dependent option pricing, Quant. Finance, 13, 6, 861-872, (2013) · Zbl 1281.91192
[42] Xu, W.; Yin, Y., Pricing American options by willow tree method under jump-diffusion process, J. Derivatives, 22, 1, 46-56, (2014)
[43] Xu, Z.; Wang, Q., Pricing guaranteed minimum withdrawal benefits: a PDE approach, In: International Conference on Management and Service Science, 1-4, (2009)
[44] Yang, S.; Dai, T., A flexible tree for evaluating guaranteed minimum withdrawal benefits under deferred life annuity contracts with various provisions, Insurance: Math. Econ., 52, 231-242, (2013) · Zbl 1284.91279
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. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.