Quantile autoregression neural network model with applications to evaluating value at risk

Abstract We develop a new quantile autoregression neural network (QARNN) model based on an artificial neural network architecture. The proposed QARNN model is flexible and can be used to explore potential nonlinear relationships among quantiles in time series data. By optimizing an approximate error function and standard gradient based optimization algorithms, QARNN outputs conditional quantile functions recursively. The utility of our new model is illustrated by Monte Carlo simulation studies and empirical analyses of three real stock indices from the Hong Kong Hang Seng Index (HSI), the US S&P500 Index (S&P500) and the Financial Times Stock Exchange 100 Index (FTSE100).

[1]  R. Engle.  Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation , 1982 .

[2]  Keming Yu,et al.  Quantile regression: applications and current research areas , 2003 .

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

[4]  Xiaohong Chen,et al.  Copula-Based Nonlinear Quantile Autoregression , 2008 .

[5]  R. Koenker.  Additive models for quantile regression: Model selection and confidence bandaids , 2010 .

[6]  J. Powell,et al.  Censored regression quantiles , 1986 .

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

[8]  Runze Li,et al.  Robust neural network with applications to credit portfolio data analysis. , 2010, Statistics and its interface.

[9]  Jianqing Fan,et al.  Quantile autoregression. Commentary , 2006 .

[10]  Sung Y. Park,et al.  Quantile Autoregressive Distributed Lag Model with an Application to House Price Returns , 2013 .

[11]  H. Nagy,et al.  Prediction of Sediment Load Concentration in Rivers using Artificial Neural Network Model , 2002 .

[12]  Wei-Min Qian,et al.  Testing in linear composite quantile regression models , 2014, Comput. Stat..

[13]  Pilar Abad,et al.  A comprehensive review of Value at Risk methodologies , 2014 .

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

[15]  T. Bollerslev,et al.  Generalized autoregressive conditional heteroskedasticity , 1986 .

[16]  Roberto Tagliaferri,et al.  A novel neural network-based survival analysis model , 2003, Neural Networks.

[17]  M. C. Jones,et al.  Local Linear Quantile Regression , 1998 .

[18]  Ghulam Ali,et al.  EGARCH, GJR-GARCH, TGARCH, AVGARCH, NGARCH, IGARCH and APARCH Models for Pathogens at Marine Recreational Sites , 2013 .

[19]  Alex J. Cannon.  Quantile regression neural networks: Implementation in R and application to precipitation downscaling , 2011, Comput. Geosci..

[20]  J. Powell,et al.  Least absolute deviations estimation for the censored regression model , 1984 .

[21]  Christian Gourieroux,et al.  Dynamic quantile models , 2008 .

[22]  Peter F. Christoffersen.  Evaluating Interval Forecasts , 1998 .

[23]  Pin T. Ng,et al.  Quantile smoothing splines , 1994 .

[24]  María Concepción Ausín,et al.  A semiparametric Bayesian approach to the analysis of financial time series with applications to value at risk estimation , 2014, Eur. J. Oper. Res..

[25]  K. I. Ramachandran,et al.  Fault diagnosis of spur bevel gear box using artificial neural network (ANN), and proximal support vector machine (PSVM) , 2010, Appl. Soft Comput..

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

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

[28]  R. Koenker,et al.  Penalized triograms: total variation regularization for bivariate smoothing , 2004 .

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

[30]  Marcelo Brutti Righi,et al.  Forecasting Value at Risk and Expected Shortfall Based on Serial Pair-Copula Constructions , 2015, Expert Syst. Appl..

[31]  C. Granger,et al.  A long memory property of stock market returns and a new model , 1993 .

[32]  Jooyoung Jeon,et al.  Using CAViaR Models with Implied Volatility for Value-at-Risk Estimation , 2013 .

[33]  David E. Allen,et al.  EVT and tail-risk modelling: Evidence from market indices and volatility series , 2013 .

[34]  Gabriel Montes-Rojas,et al.  Threshold quantile autoregressive models , 2011 .

[35]  Marc G. Genton,et al.  Local Polynomial Quantile Regression With Parametric Features , 2009 .

[36]  I Cheng Yeh.  Estimating Distribution of Concrete Strength Using Quantile Regression Neural Networks , 2014 .

[37]  Fedya Telmoudi,et al.  Consistent Estimation of the Value at Risk When the Error Distribution of the Volatility Model is Misspecified , 2016 .

[38]  Chang Shu,et al.  Artificial neural network ensembles and their application in pooled flood frequency analysis , 2004 .

[39]  A. McNeil,et al.  Estimation of tail-related risk measures for heteroscedastic financial time series: an extreme value approach , 2000 .

[40]  Gregory Kordas.  Smoothed binary regression quantiles , 2006 .