An exponentially weighted quantile regression via SVM with application to estimating multiperiod VaR

The square root of time rule under RiskMetrics has been used as an important tool to estimate multiperiod value at risk (VaR). However, the conditions for the rule are too restrictive to get empirical support in practice since multiperiod VaR is a complex nonlinear function of the holding period and the one-step ahead volatility forecast. In this paper, we propose a new model by considering an exponentially weighted quantile regression via SVM to provide greater accuracy for multiperiod VaR measure. In both numerical simulations and empirical studies on three stock indices, the proposed model outperforms several traditional methods including the volatility models, filtered historical simulation, and linear quantile regression approaches in terms of the value of the number of significant entries, the mean absolute error, and the p value of prediction test in Harvey et al. (Int J Forecast 13:281–291, 1997).

[1]  R. Koenker,et al.  Robust Tests for Heteroscedasticity Based on Regression Quantiles , 1982 .

[2]  R. Koenker,et al.  Regression Quantiles , 2007 .

[3]  Shouyang Wang,et al.  Multiple-v support vector regression based on spectral risk measure minimization , 2013, Neurocomputing.

[4]  Ying Chen,et al.  Associating stock prices with web financial information time series based on support vector regression , 2013, Neurocomputing.

[5]  Ji Zhu,et al.  Quantile Regression in Reproducing Kernel Hilbert Spaces , 2007 .

[6]  Qifa Xu,et al.  Weighted quantile regression via support vector machine , 2015, Expert Syst. Appl..

[7]  Peter Christoffersen,et al.  Série Scientifique Scientific Series 2003 s-05 Backtesting Value-at-Risk : A Duration-Based Approach , 2003 .

[8]  F. Diebold,et al.  Comparing Predictive Accuracy , 1994, Business Cycles.

[9]  Jing Wang,et al.  Bayesian quantile regression for parametric nonlinear mixed effects models , 2012, Stat. Methods Appl..

[10]  Daniel R. Smith,et al.  Evaluating Value-at-Risk Models via Quantile Regression , 2008 .

[11]  Graeme Laurie.  An End to Error of Law in Unjust Enrichment: Morgan Guaranty Trust Company of New York v. Lothian Regional Council , 1995 .

[12]  Phhilippe Jorion.  Value at Risk: The New Benchmark for Managing Financial Risk , 2000 .

[13]  Xinyuan Song,et al.  Oracle model selection for nonlinear models based on weighted composite quantile regression accelerated failure time model , 2012 .

[14]  Charlene Xie,et al.  Generalized Autoregressive Conditional Heteroskedasticity in Credit Risk Measurement , 2009, 2009 International Conference on Management and Service Science.

[15]  U. Cherubini,et al.  RiskMetrics Technical Document , 2015 .

[16]  Giovanni Barone-Adesi,et al.  VaR without correlations for portfolios of derivative securities , 1999 .

[17]  Paul Newbold,et al.  Testing the equality of prediction mean squared errors , 1997 .

[18]  Alexander J. Smola,et al.  Nonparametric Quantile Estimation , 2006, J. Mach. Learn. Res..

[19]  Ming Yuan,et al.  GACV for quantile smoothing splines , 2006, Comput. Stat. Data Anal..

[20]  Tae-Hwan Kim,et al.  VAR for VaR: Measuring Tail Dependence Using Multivariate Regression Quantiles , 2015, SSRN Electronic Journal.

[21]  G. Barone-Adesi.  VaR Without Correlations for Nonlinear Portfolios , 1998 .

[22]  James W. Taylor.  Using Exponentially Weighted Quantile Regression to Estimate Value at Risk and Expected Shortfall , 2008 .

[23]  Mike K. P. So,et al.  On conditional moments of garch models, with applications to multiple period value at risk estimation , 2003 .

[24]  Michael McAleer,et al.  Forecasting Value-at-Risk Using Nonlinear Regression Quantiles and the Intra-day Range , 2011 .

[25]  Sophie Dabo-Niang,et al.  Consistency of a nonparametric conditional mode estimator for random fields , 2014, Stat. Methods Appl..

[26]  Wagner Piazza.  Evaluating Value-at-Risk models via Quantile Regression , 2009 .

[27]  James W. Taylor.  A Quantile Regression Approach to Estimating the Distribution of Multiperiod Returns , 1999 .

[28]  Fulvio Corsi,et al.  A Simple Approximate Long-Memory Model of Realized Volatility , 2008 .

[29]  James W. Taylor.  A Quantile Regression Neural Network Approach to Estimating the Conditional Density of Multiperiod Returns , 2000 .

[30]  Changha Hwang,et al.  Support Vector Quantile Regression with Weighted Quadratic Loss Function , 2010 .

[31]  Jooyong Shim,et al.  Estimating value at risk with semiparametric support vector quantile regression , 2012, Comput. Stat..

[32]  D. Hunter.  Incorporating volatility updating into the historical simulation method for value-at-risk , 1998 .

[33]  Wing Lon Ng,et al.  Nonparametric estimation of general multivariate tail dependence and applications to financial time series , 2015, Stat. Methods Appl..

[34]  Akiko Takeda.  Generalization performance of nu-support vector classifier based on conditional value-at-risk minimization , 2009, Neurocomputing.

[35]  R. Koenker.  Quantile Regression: Name Index , 2005 .

[36]  Paul H. Kupiec,et al.  Techniques for Verifying the Accuracy of Risk Measurement Models , 1995 .