Abstract
Risk management has attracted a great deal of attention, and Value at Risk (VaR) has emerged as a particularly popular and important measure for detecting the market risk of financial assets. The quantile regression method can generate VaR estimates without distributional assumptions; however, empirical evidence has shown the approach to be ineffective at evaluating the real level of downside risk in out-of-sample examination. This paper proposes a process in VaR estimation with methods of quantile regression and kernel estimator which applies the nonparametric technique with extreme quantile forecasts to realize a tail distribution and locate the VaR estimates. Empirical application of worldwide stock indices with 29 years of data is conducted and confirms the proposed approach outperforms others and provides highly reliable estimates.
Similar content being viewed by others
Notes
In 1994, JP Morgan published its risk-measured program RiskMetrics which, for the first time, systematically developed detailed methodologies for VaR. In 1996, the Basel committee on banking supervision amended the Basel Capital Accord and obliged their member banks to reserve capital requirements calculated based on VaR.
Baixauli and Alvarez (2006) showed that accurate VaR estimates can be produced with correct characterization of a left-tail distribution.
In empirical applications of this study, the process is applied rolling through the entire out-of-sample, so the tail distributions are generated for each day and are different from each other. The demonstration in the figure is done as an illustration from two selected dates: August 31, 2009 (the last sampling day) and September 1, 2007 (2 years prior to the last date).
Both settings are empirically rational choices rather than selections according to statistical inferences. Thousandth quantile forecasts in empirical applications are considered very frequent and the threshold of 0.02 is reasonably extreme for determining the 1 % unconditional quantile.
The two indices are extracted with different sample periods due to data availability and stability.
Samples of emerging markets also end at August 31, 2009.
Both VaR estimate series are transformed into percentage format to coincide with return series in same scale.
The only exception is that the QR-symmetric model also has a favorable unconditional likelihood ratio test at a 10 % significance level for the FTSE100.
With only one exception where both QR-asymmetric and KQ-asymmetric models have the same back-testing outcome of 38 for the Russian stock index with out-of-samples of 2,858.
References
Angelidis T, Benos A, Degiannakis S (2007) A robust VaR model under different time periods and weighting schemes. Rev Quant Finan Acc 28:187–201
Baillie RT, Bollerslev T, Mikkelsen MO (1996) Fractionally integrated generalized autoregressive conditional heteroskedasticity. J Econom 74:3–30
Baixauli JS, Alvarez S (2006) Evaluating effects if excess kurtosis on VaR estimates: evidence for international stock indices. Rev Quant Finan Acc 27:27–46
Bali TG, Mo H, Tang Y (2008) The role of autoregressive conditional skewness and kurtosis in the estimation of conditional VaR. J Bank Finance 32:269–282
Bollerslev T (1986) Generalized autoregressive conditional heteroscedasticity. J Econom 31:307–327
Brooks C, Burke SP, Persand G (2005) Autoregressive conditional kurtosis. J Financ Econ 3:399–421
Butler JS, Schachter B (1997) Estimating value-at-risk with a precision measure by combining kernel estimation with historical simulation. Rev Deriv Res 1:371–390
Cai Z, Wang X (2008) Nonparametric estimation of conditional VaR and expected shortfall. J Econom 147:120–130
Chen H (2008) Value-at-Risk efficient portfolio selection using goal programming. Rev Pacific Basin Financ Markets Policies 11(2):187–200
Chen FY, Liao SL (2009) Modelling VaR for foreign-asset portfolios in continuous time. Econ Model 26:234–240
Chou RY, Wu CC, Liu N (2009) Forecasting time-varying covariance with a range-based dynamic conditional correlation model. Rev Quant Finan Acc 33:327–345
Christoffersen PF (1998) Evaluating interval forecast. Int Econ Rev 39:841–862
Connor G, Linton O (2007) Semiparametric estimation of a characteristic-based factor model of common stock returns. J Empir Finance 14:694–717
Costello A, Asem E, Gardner E (2008) Comparison of historically simulated VaR: evidence from oil prices. Energy Econ 30:2154–2166
Danielsson J, de Vries CG (2000) Value at risk and extreme return. Ann Econ Stat 60:239–270
Engle RF, Manganelli S (2004) CAViaR: conditional autoregressive value at risk by regression quantiles. J Bus Econ Stat 22:367–381
Epanechnikov VA (1969) Nonparametric estimation of a multivariate probability density. Theo Prob Its App 14:153–158
Gijbels I, Pope A, Wand MP (1999) Understanding exponential smoothing via kernel regression. J R Stat Soc Series B 61:39–50
Giot P, Laurent S (2004) Modelling daily Value-at-Risk using realized volatility and ARCH type models. J Empir Finance 11:379–398
Glosten LR, Jagannathan R, Runkle DE (1993) On the relation between the expected value and the volatility of the normal excess return on stocks. J Financ 48:1779–1801
Gourieroux C, Laurent JP, Scaillet O (2000) Sensitivity analysis of Value at Risk. J Empir Finance 7:225–245
Gray SF (1996) Modeling the conditional distribution of interest rates as a regime-switching process. J Financ Econ 42:27–62
Harvey CR, Siddique A (1999) Autoregressive conditional skewness. J Financ Quant Anal 34:465–487
Hendricks D (1996) Evaluating value-at-risk models using historical data. Federal Reserve Bank New York Econ Policy Rev, pp 39–69
Hsu CP, Huang CW, Chiou WJ (2011) Effectiveness of copula-extreme value theory in estimating value-at-risk: empirical evidence from Asian emerging markets. Rev Quant Finan Acc. doi:10.1007/s11156-011-0261-0
Huang AY (2009) A value-at-risk approach with kernel estimator. Appl Financ Econ 19:379–395
Huang AY (2011) Volatility modeling by asymmetrical quadratic effect with diminishing marginal impact. Comp Econ 37:301–330
Jondeau E, Rockinger M (2003) Conditional volatility, skewness, and kurtosis: existence, persistence, and comovements. J Econ Dyn Control 27:1699–1737
Jones MC, Marron JS, Sheather SJ (1996) A brief survey of bandwidth selection for density estimation. J Am Stat Assoc 91:401–407
Jorion P (2006) Value at risk: the new benchmark for managing financial risk, 3rd edn. McGraw-Hill
Kahneman I, Tversky A (1979) Prospect theory: an analysis of decision under risk. Econometrica 47:263–290
Koenker R, Basset G (1978) Regression quantiles. Econometrica 46:33–50
Kuan CM, Yeh JH, Hsu YC (2009) Assessing value at risk with CARE, the conditional autoregressive expectile models. J Econom 150:261–270
Kuester K, Mittnik S, Paolella MS (2006) Value-at-Risk prediction: a comparison of alternative strategies. J Financ Econom 4:53–89
Ledoit O, Wolf M (2008) Robust performance hypothesis testing with the Sharpe ratio. J Empir Finance 15:850–859
Lee CF, Su JB (2011) Alternative statistical distributions for estimating value-at-risk: theory and evidence. Rev Quant Finan Acc. doi:10.1007/s11156-011-0256-x
Lu C, Tse Y, Williams M (2012) Returns transmission, value at risk, and diversification benefits in international REITs: evidence from the financial crisis. Rev Quant Finan Acc. doi:10.1007/s11156-012-0274-3
Markowitz H (1952) Portfolio selection. J Financ 7:77–91
McNeil AJ, Frey R (2000) Estimation of tail-related risk measures for heteroscedastic financial time series: an extreme value approach. J Empir Finance 7:71–300
Mittnik S, Paolella MS, Rachev S (2002) Stationarity of stable power-GARCH processes. J Econom 106:97–107
Neftci SN (2000) Value at Risk calculations, extreme events, and tail estimation. J Deriv 7:23–37
Nelson DB (1991) Conditional heteroskedasticity in asset returns: a new approach. Econometrica 59:347–370
Poon S, Granger CW (2003) Forecasting volatility in financial markets: a review. J Econ Lit 41:478–539
Sentana E (1995) Quadratic ARCH models. Rev Econ Stud 62:639–661
Sheather SJ, Marron JS (1990) Kernel quantile estimators. J Am Stat Assoc 85:410–416
Silverman BW (1986) Density estimation for statistics and data analysis. Chapman and Hall, London
Taylor JW (2008) Using exponentially weighted quantile regression to estimate value at risk and expected shortfall. J Financ Econom 6:382–406
Venkataraman S (1997) Value at risk for a mixture of normal distributions: the use of quasi-Bayesian estimation techniques. Econ Perspect 21:2–13
Acknowledgments
This research is supported in part by the National Science Council of Taiwan (NSC100-2632-H-155-001-MY2). The author thanks Wei-Hwa Peng and Li-Chiang Chen for their excellent research assistance.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Huang, A.Y. Value at risk estimation by quantile regression and kernel estimator. Rev Quant Finan Acc 41, 225–251 (2013). https://doi.org/10.1007/s11156-012-0308-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11156-012-0308-x