Scenario aggregation method for portfolio expectile optimization
-
Edgars Jakobsons
Abstract
The statistical functional expectile has recently attracted the attention of researchers in the area of risk management, because it is the only risk measure that is both coherent and elicitable. In this article, we consider the portfolio optimization problem with an expectile objective. Portfolio optimization problems corresponding to other risk measures are often solved by formulating a linear program (LP) that is based on a sample of asset returns. We derive three different LP formulations for the portfolio expectile optimization problem, which can be considered as counterparts to the LP formulations for the Conditional Value-at-Risk (CVaR) objective in the works of Rockafellar and Uryasev [43], Ogryczak and Śliwiński [41] and Espinoza and Moreno [21]. When the LPs are based on a simulated sample of the true (assumed continuous) asset returns distribution, the portfolios obtained from the LPs are only approximately optimal. We conduct a numerical case study estimating the suboptimality of the approximate portfolios depending on the sample size, number of assets, and tail-heaviness of the asset returns distribution. Further, the computation times using the three LP formulations are analyzed, showing that the formulation that is based on a scenario aggregation approach is considerably faster than the two alternatives.
Funding statement: The author gratefully acknowledges financial support from RiskLab Switzerland and the Swiss Finance Institute.
Acknowledgements
This work has benefited from stimulating discussions with Fabio Bellini and Christian Colombo during a visit to the University of Milano–Bicocca in January 2016.
References
[1] Aas K. and Hobæk Haff I., The generalized hyperbolic skew Student’s t-distribution, J. Financ. Econ. 4 (2006), no. 2, 275–309. 10.1093/jjfinec/nbj006Search in Google Scholar
[2] Acerbi C. and Szekely B., Back-testing expected shortfall, Risk 27 (2014), no. 11, 76–81. Search in Google Scholar
[3] Ang J. S., A note on the E, SL portfolio selection model, J. Financ. Quant. Anal. 10 (1975), no. 05, 849–857. 10.2307/2330276Search in Google Scholar
[4] Artzner P., Delbaen F., Eber J.-M. and Heath D., Coherent measures of risk, Math. Finance 9 (1999), no. 3, 203–228. 10.1017/CBO9780511615337.007Search in Google Scholar
[5] Barndorff-Nielsen O. E., Exponentially decreasing distributions for the logarithm of particle size, Proc. Roy. Soc. Lond. A. Math. Phys. Eng. Sci. 353 (1977), no. 1674, 401–419. 10.1098/rspa.1977.0041Search in Google Scholar
[6] Barndorff-Nielsen O. E., Normal inverse Gaussian distributions and stochastic volatility modelling, Scand. J. Stat. 24 (1997), no. 1, 1–13. 10.1111/1467-9469.00045Search in Google Scholar
[7] Bellini F. and Bignozzi V., On elicitable risk measures, Quant. Finance 15 (2015), no. 5, 725–733. 10.1080/14697688.2014.946955Search in Google Scholar
[8] Bellini F., Colombo C. and Pınar M. Ç., Portfolio optimization with expectiles, Poster presented at the workshop Dependence & Risk Measures, University of Milano–Bicocca, November 2015. Search in Google Scholar
[9] Bellini F. and Di Bernardino E., Risk management with expectiles, European J. Finance (2015), 10.1080/1351847X.2015.1052150. 10.1080/1351847X.2015.1052150Search in Google Scholar
[10] Bellini F., Klar B., Müller A. and Gianin E. R., Generalized quantiles as risk measures, Insurance Math. Econom. 54 (2014), 41–48. 10.1016/j.insmatheco.2013.10.015Search in Google Scholar
[11] Bernardo A. E. and Ledoit O., Gain, loss, and asset pricing, J. Political Econ. 108 (2000), no. 1, 144–172. 10.1086/262114Search in Google Scholar
[12] Bixby R. E., Solving real-world linear programs: A decade and more of progress, Oper. Res. 50 (2002), no. 1, 3–15. 10.1287/opre.50.1.3.17780Search in Google Scholar
[13] Blattberg R. C. and Gonedes N. J., A comparison of the stable and Student distributions as statistical models for stock prices, J. Bus. 47 (1974), no. 2, 244–280. 10.1142/9789814287067_0003Search in Google Scholar
[14] Carver L., Mooted VaR substitute cannot be back-tested, says top quant, Risk 2013 (2013). Search in Google Scholar
[15] Carver L., Back-testing expected shortfall: Mission impossible?, Risk 2014 (2014). Search in Google Scholar
[16] Delbaen F., Monetary Utility Functions, Osaka University Press, Osaka, 2012. Search in Google Scholar
[17] Delbaen F., A remark on the structure of expectiles, preprint 2013, http://arxiv.org/abs/1307.5881. Search in Google Scholar
[18] Delbaen F., Bellini F., Bignozzi V. and Ziegel J. F., Risk measures with the CxLS property, Finance Stoch. 20 (2016), no. 2, 433–453. 10.1007/s00780-015-0279-6Search in Google Scholar
[19] De Rossi G. and Harvey A., Quantiles, expectiles and splines, J. Econometrics 152 (2009), no. 2, 179–185. 10.1016/j.jeconom.2009.01.001Search in Google Scholar
[20] Efron B., Regression percentiles using asymmetric squared error loss, Statist. Sinica 1 (1991), no. 1, 93–125. Search in Google Scholar
[21] Espinoza D. and Moreno E., A primal-dual aggregation algorithm for minimizing conditional-value-at-risk in linear programs, Comput. Optim. Appl. 59 (2014), no. 3, 617–638. 10.1007/s10589-014-9692-6Search in Google Scholar
[22] Föllmer H. and Schied A., Stochastic Finance: An Introduction in Discrete Time, Walter de Gruyter, Berlin, 2004. 10.1515/9783110212075Search in Google Scholar
[23] Gneiting T., Making and evaluating point forecasts, J. Amer. Statist. Assoc. 106 (2011), no. 494, 746–762. 10.1198/jasa.2011.r10138Search in Google Scholar
[24] Gondzio J., Interior point methods 25 years later, European J. Oper. Res. 218 (2012), no. 3, 587–601. 10.1016/j.ejor.2011.09.017Search in Google Scholar
[25] Hull J. and White A., Hull and White on the pros and cons of expected shortfall, Risk 2014 (2014). Search in Google Scholar
[26] Jakobsons E., Suboptimality in portfolio conditional value-at-risk optimization, J. Risk 18 (2016), no. 4, 1–23. 10.21314/J0R.2016.330Search in Google Scholar
[27] Jakobsons E. and Vanduffel S., Dependence uncertainty bounds for the expectile of a portfolio, Risks 3 (2015), no. 4, 599–623. 10.3390/risks3040599Search in Google Scholar
[28] Jorion P., Value at Risk: The New Benchmark for Managing Financial Risk, McGraw–Hill, New York, 1996. Search in Google Scholar
[29] Kapsos M., Zymler S., Christofides N. and Rustem B., Optimizing the Omega ratio using linear programming, J. Comput. Finance 17 (2014), no. 4, 49–57. 10.21314/JCF.2014.283Search in Google Scholar
[30] Keating C. and Shadwick W. F., A universal performance measure, J. Perform. Measurement 6 (2002), no. 3, 59–84. Search in Google Scholar
[31] Konno H. and Yamazaki H., Mean-absolute deviation portfolio optimization model and its applications to Tokyo stock market, Manag. Sci. 37 (1991), no. 5, 519–531. 10.1287/mnsc.37.5.519Search in Google Scholar
[32] Kuan C.-M., Yeh J.-H. and Hsu Y.-C., Assessing value at risk with CARE, the conditional autoregressive expectile models, J. Econometrics 150 (2009), no. 2, 261–270. 10.1016/j.jeconom.2008.12.002Search in Google Scholar
[33] Madan D. B. and Seneta E., The variance gamma (VG) model for share market returns, J. Bus. 63 (1990), no. 4, 511–524. 10.1086/296519Search in Google Scholar
[34] Mansini R., Ogryczak W. and Speranza M. G., LP solvable models for portfolio optimization: A classification and computational comparison, IMA J. Manag. Math. 14 (2003), no. 3, 187–220.10.1093/imaman/14.3.187Search in Google Scholar
[35] Mansini R., Ogryczak W. and Speranza M. G., Conditional value at risk and related linear programming models for portfolio optimization, Ann. Oper. Res. 152 (2007), no. 1, 227–256. 10.1007/s10479-006-0142-4Search in Google Scholar
[36] Mansini R., Ogryczak W. and Speranza M. G., Twenty years of linear programming based portfolio optimization, European J. Oper. Res. 234 (2014), no. 2, 518–535. 10.1016/j.ejor.2013.08.035Search in Google Scholar
[37] Mansini R., Ogryczak W. and Speranza M. G., Linear and Mixed Integer Programming for Portfolio Optimization, Springer, Cham, 2015. 10.1007/978-3-319-18482-1Search in Google Scholar
[38] Markowitz H., Portfolio selection, J. Finan. 7 (1952), no. 1, 77–91. 10.12987/9780300191677Search in Google Scholar
[39] McNeil A. J., Frey R. and Embrechts P., Quantitative Risk Management: Concepts, Techniques and Tools, Princeton University Press, Princeton, 2005. Search in Google Scholar
[40] Newey W. K. and Powell J. L., Asymmetric least squares estimation and testing, Econometrica 55 (1987), no. 4, 819–847. 10.2307/1911031Search in Google Scholar
[41] Ogryczak W. and Śliwiński T., On solving the dual for portfolio selection by optimizing conditional value at risk, Comput. Optim. Appl. 50 (2011), no. 3, 591–595. 10.1007/s10589-010-9321-ySearch in Google Scholar
[42] Rémillard B., Statistical Methods for Financial Engineering, CRC Press, Boca Raton, 2013. Search in Google Scholar
[43] Rockafellar R. T. and Uryasev S., Optimization of conditional value-at-risk, J. Risk 2 (2000), no. 3, 21–42. 10.21314/JOR.2000.038Search in Google Scholar
[44] Sharpe W. F., A linear programming approximation for the general portfolio analysis problem, J. Financ. Quant. Anal. 6 (1971), no. 5, 1263–1275. 10.2307/2329860Search in Google Scholar
[45] Yao Q. and Tong H., Asymmetric least squares regression estimation: A nonparametric approach, J. Nonparametr. Stat. 6 (1996), no. 2–3, 273–292. 10.1080/10485259608832675Search in Google Scholar
[46] Ziegel J. F., Coherence and elicitability, Mathematical Finance (2014), 10.1111/mafi.12080. 10.1111/mafi.12080Search in Google Scholar
[47] Basel Committee on Banking Supervision, Fundamental review of the trading book, 2012, available at www.bis.org/publ/bcbs219.htm. Search in Google Scholar
© 2016 by De Gruyter
