Estimating value at risk with semiparametric support vector quantile regression Authors Jooyong Shim Department of Data Science and Institute of Statistical Information Inje University Yongtae Kim Department of Statistics Dankook University Jangtaek Lee Department of Statistics Dankook University Changha Hwang Department of Statistics Dankook University Original Paper

First Online: 09 October 2011 Received: 07 January 2011 Accepted: 23 September 2011 DOI :
10.1007/s00180-011-0283-z

Cite this article as: Shim, J., Kim, Y., Lee, J. et al. Comput Stat (2012) 27: 685. doi:10.1007/s00180-011-0283-z
Abstract Value at Risk (VaR) has been used as an important tool to measure the market risk under normal market. Usually the VaR of log returns is calculated by assuming a normal distribution. However, log returns are frequently found not normally distributed. This paper proposes the estimation approach of VaR using semiparametric support vector quantile regression (SSVQR) models which are functions of the one-step-ahead volatility forecast and the length of the holding period, and can be used regardless of the distribution. We find that the proposed models perform better overall than the variance-covariance and linear quantile regression approaches for return data on S&P 500, NIKEI 225 and KOSPI 200 indices.

Keywords EWMA GARCH t -GARCHQuantile regression Semiparametric support vector quantile regression Value at risk

References Alexander CO, Leigh CT (1997) On the covariances matrices used in value at risk models. J Deriv 4: 50–62

CrossRef Bollerslev T (1986) Generalized autoregressive conditional heteroskedasticity. J Econom 31: 307–327

MathSciNet MATH CrossRef Bollerslev T (1987) A conditional heteroskedastic time series model for speculative prices and rates of returns. Rev Econom Stat 69: 542–547

CrossRef Boudoukh J, Richardson M, Whitelaw RF (1997) Investigation of a class of volatility estimators. J Deriv 4: 63–71

CrossRef Chen MJ, Chen J (2002) Application of quantile regression to estimation of value at risk. Working paper, National Chung-Cheng University

Diebold FX, Hickman A, Inoue A, Schuermann T (1998) Scale models. Risk 11: 104–107

Engle R, Manganelli S (2004) CAViaR: conditional autoregressive value at risk by regression quantiles. J Bus Econom Stat 22: 367–381

MathSciNet CrossRef Fan J, Gijbels I (1996) Local polynomial modelling and its applications. Chapman and Hall, London

MATH Glasserman P, Heidelberger P, Shahabuddin P (2002) Portfolio value-at-risk with heavy-tailed risk factors. Math Finance 12: 239–269

MathSciNet MATH CrossRef Härdle W (1990) Applied nonparametric regression. Cambridge University Press, New York

MATH Hull J, White A (1998) Value at risk when daily changes in market variables are not normally distributed. J Deriv 5: 9–19

CrossRef Jorion P (2007) Value at risk: the new benchmark for managing financial risk. McGraw-Hill, New York

Kohenker R, Bassett G (1978) Regression quantiles. Econometrica 46: 33–50

MathSciNet CrossRef Koenker R, Ng P, Portnoy S (1994) Quantile smoothing splines. Biometrika 81: 673–680

MathSciNet MATH CrossRef Kroner KF, Kneafsey KP, Claessen S (1995) Forecasting volatility in common markets. J Forecast 14: 77–95

CrossRef Kuester K, Mittnik S, Paolella MS (2006) Value-at-risk prediction: a comparison of alternative strategies. J Financial Econom 4: 53–89

Kuhn HW, Tucker AW (1951) Nonlinear programming. In: Proceedings of 2nd Berkeley symposium. University of California Press, Berkeley, pp 481–492

Li Y, Liu Y, Zhu J (2007) Quantile regression in reproducing kernel Hilbert spaces. J Am Stat Assoc 102: 255–268

MathSciNet MATH CrossRef Nychka D, Gray G, Haaland P, Martin D, O’Connell M (1995) A nonparametric regression approach to syringe grading for quality improvement. J Am Stat Assoc 90: 1171–1178

MATH CrossRef Shim J, Hwang C (2009) Support vector censored quantile regression under random censoring. Comput Stat Data Anal 53: 912–919

MathSciNet MATH CrossRef Smola AJ, Frieβ TT, Schölkopf B (1999) Semiparametric support vector and linear programming machines. Adv Neural Inf Process Syst 11: 585–591

Takeuchi I, Furuhashi T (2004) Non-crossing quantile regressions by SVM. In: Proceedings of 2004 IEEE international joint conference on neural networks, pp 401–406

Taylor JW (1999) A quantile regression approach to estimating the distribution of multiperiod returns. J Deriv 7: 64–78

CrossRef Taylor JW (2000) A quantile regression neural network approach to estimating the conditional density of multiperiod returns. J Forecast 19: 299–311

CrossRef Vapnik VN (1995) The nature of statistical learning theory. Springer, New York

MATH White H (1992) Nonparametric estimation of conditional quantiles using neural networks. In: Artificial neural networks: Approximation and learning theory. Blackwell, Oxford, pp 191–205

Yuan M (2006) GACV for quantile smoothing splines. Comput Stat Data Anal 50: 813–829

MATH CrossRef Zakai A, Ritov Y (2009) Consistency and localizability. J Mach Learn Res 10: 827–856

MathSciNet MATH