摘要
商品期货收益率序列存在明显的厚尾现象,将扰动项设定为厚尾分布的波动率模型要优于普通Gaussian分布以及t分布模型。以7种损失函数作为评价准则,在扰动项分别服从偏t分布以及广义双曲线偏t分布时比较了波动率RV、二次幂变差BV以及medRV3种不同已实现测度下Realized GARCH模型对商品期货的波动性预测能力。最后得到扰动项服从ghst分布的Realized GARCH模型具有更优的预测能力。
There is obvious fat tail phenomenon in the yield series of commodity futures,and the volatility model with the fat tailed distribution is better than the ordinary Gaussian distribution and the T distribution model. This paper,using seven kinds of loss function as the evaluation criterion,compared the predictive ability of three realized measures of realized variance,bipower variation and medRV based on skewed T distribution and ghst distribution. Finally,under the three different realized measures,the Realized GARCH model with disturbance subject to the ghst distribution has better prediction ability than skewed T distribution.
引文
[1]Hansen P R,Huang Z,Shek H H.Realized GARCH:a joint model for returns and realized measures of volatility[J].Journal of Applied Econometrics,2012,27(06):877-906.
[2]Tian S,Hamori S.Modeling interest rate volatility:a realized GARCH approach[J].Journal of Banking and Finance,2015,61:158-171.
[3]Louzis D P,Xanthopoulos-Sisinis S,Refenes A P.Realized volatility models and alternative value at risk prediction strategies[J].Economic Modelling,2014,40:101-116.
[4]Takeuchinogimori A.An empirical analysis of the Nikkei225 put options using realized GARCH models[J].Studies in Health Technology and Informatics,2012,180(01):143-147.
[5]黄友珀,唐振鹏,周熙雯.基于偏t分布realized GARCH模型的尾部风险估计[J].系统工程理论与实践,2015,35(09):2200-2208.
[6]刘若萌,郭名媛.基于厚尾分布下Realized GARCH模型的中国股票市场波动研究[J].天津理工大学学报,2017,33(05):46-50.
[7]魏正元,余徳英,李素平.已实现GARCH-GED模型的研究及应用[J].重庆理工大学学报,2018(3):273-278.
[8]田凤平,杨科.基于TVS-HAR模型的农产品期货市场已实现波动率的预测研究[J].系统工程理论与实践,2016,36(12):3003-3016.
[9]Brandt M W,Jones C S,Volatility forecasting with range-based EGARCH models[J].Journal of Business and Economic Statistics,2006,24(4):470-486.
[10]Engle R F,Gallo G M.A multiple indicators model for volatility using intra-daily data[J].Journal of Econometrics,2006,131(1):3-27.
[11]Ghysels E,Santa-Clara P,Valkanov R.Predicting volatility:Getting the most out of return data sampled at different frequencies[J].Econometric Reviews,2007,26(1):53-90.
[12]Ghysels E,Sinko A,Valkanov R.MIDAS regressions:Further results and new directions[J].Journal of Econometrics,2006,131(1):59-95.
[13]Barndorff-Nielsen O E,Shephard N.Power and bipower variation with stochastic volatility and jumps[J].Journal of Financial Econometrics,2004,2(1):1-37.
[14]王天一,黄卓.高频数据波动率建模:基于厚尾分布的Realized GARCH模型[J].数量经济技术经济研究,2012(05):149-160.
[15]Corsi F,Pirino D,Reno R.Threshold bipower variation and the impact of jumps on volatility forecasting[J].Journal of Econometrics,2010,159(2):276-288.
[16]Barndorffnielsen O E,Shephard N,Power and bipower variation with stochastic volatility and jumps[J].Econometric Papers,2003,2(2003-W17):1-37.
[17]Andersen T G,Dobrev D,Schaumburg E.Jump-robust volatility estimation using nearest neighbor truncation[J].Journal of Econometrics,2012,169(01):75-93.
[18]Bollerslev T P.A conditional heteroskedastic time series model for speculative price and rates of return[J].Review of Economics and Statistics,1987,69(03):542-547.
[19]Fernandez C,Steel M F J.On Bayesian modeling of fat tails and skewness[J].Journal of the American Statistical Association,1998,93(441):359-371.
[20]Aas K,Haff I H.The generalized hyperbolic skew student’s t-distribution[J].Social Science Electronic Publishing,2006,4(02):275-309.
[21]Hansen B E.Autoregressive conditional density estimation[J].International Economic Review,1994,35(03):705-730.