基于马尔科夫状态转换和跳跃的高频波动率模型预测被引量：16

The Forecasting Performance of the High-frequency Volatility Models with the Markov-switching Regime and Jump

导出

摘要以HAR-RV、HAR-RV-J、HAR-RV-CJ和HAR-RV-TCJ模型为基础,结合马尔科夫状态转换机制,构建了四种新的马尔科夫状态转换高频波动率模型。同时,用沪深300股指期货的高频数据,运用滚动时间窗的样本外预测方法和信度设定检验(Model Confidence Set,MCS),分析了四种新模型对未来波动率的预测能力。实证结果表明,总体上,四种新的马尔科夫状态转换模型比原有波动率模型具有更好的预测表现;在众多波动率模型里,MS-HAR-RV-TCJ模型具有更高的预测精度。然而,实务界常用的低频波动率模型(如GARCH等)的预测能力表现得并不突出。 Based on the four basic high-frequency models：HAR-RV,HAR-RV-J,HAR-RV-CJ and HAR-RV-TCJ and combined the Markov-switching regime,we firstly propose four new Markov-switching regime high-frequency volatility models.Taking 5-minute high frequency data of the CSI 300 index futures contracts for example,we apply the out-of-sample rolling time window forecasting combined with Model Confidence Set which is proved superior to SPA test,to explore the forecasting performance of the new models.The empirical results show that the Markov-switching regime models have better performance in forecasting in all.Moreover,the MS-HAR-RV-TCJ model is the best model among models we have discussed in this paper.However,the GARCH-types which are popular in financial academe and practice,perform worst for volatility predicting of CSI300 index futures.

作者马锋魏宇黄登仕夏泽安

机构地区西南交通大学经济管理学院西南财经大学金融智能与金融工程四川省重点实验室

出处《系统工程》 CSSCI CSCD 北大核心 2016年第1期10-16,共7页 Systems Engineering

基金国家自然科学基金资助项目(71372109 71371157 71401077) 高等学校博士学科点专项科研基金资助课题(20120184110020) 四川省科技青年基金资助项目(15QNJJ0032) 西南交通大学研究生创新实验实践项目(YC201405118)

关键词马尔科夫状态转换机制 HAR-RV 跳跃 MCS检验 Markov-switching Regime HAR-RV Models Jump MCS Test

分类号 F830 [经济管理—金融学]

引文网络
相关文献

参考文献24

1Andersen T G,Bollerslev T. Answering the skeptics.. Yes, standard volatility models do provide accurate forecasts EJT. International Economic Review,1998, 39(4) :885-905.
2Andersen T G, et al. The distribution of realized stock return volatility l-J-. Journal of Financial Economics, 2001,61 (1) : 43- 76.
3Corsi F. A simple approximate long-memory model of realized volatility [J]. Journal of Financial Econometrics, 2009,7(2) : 174- 196.
4Granger C W J, Ding Z. Varieties of long memory models[J]. Journal of Econometrics, 1996,73(1) ..61 -77.
5Longin F M. The threshold effect in expected volatility: A model based on asymmetric information [J]. Review of Financial Studies, 1997, 10(3): 837 -869.
6Raggi D, et al. Long memory and nonlinearities in realized volatility: A markov switching approach I-J 1. Computational Statistics - Data Analysis, 2012,56 (11) : 3730- 3742.
7孙金丽,张世英.具有结构转换的GARCH模型及其在中国股市中的应用[J].系统工程,2003,21(6):86-91. 被引量：29
8Lee S S,Mykland P A. Jumps in financial markets: A new nonparametric test and jump dynamics [J]. Review of Financial Studies, 21 (6) : 2535-2563.
9Barndorff-Nielsen O E, Shephard N. Power and bipower variation with stochastic volatility and jumps -J]. Journal of Financial Econometrics, 2004,2(1): 1-48.
10Andersen T G, Bollerslev T, Diebold F X. Roughing it up: Including jump components in the measurement, modeling and forecasting of return volatility -JT. The Review of Economics and Statistics, 2007,89 (4) : 701- 720.

二级参考文献49

1徐正国,张世英.调整"已实现"波动率与GARCH及SV模型对波动的预测能力的比较研究[J].系统工程,2004,22(8):60-63. 被引量：51
2魏宇,黄登仕.基于多标度分形理论的金融风险测度指标研究[J].管理科学学报,2005,8(4):50-59. 被引量：39
3徐正国,张世英.多维高频数据的“已实现”波动建模研究[J].系统工程学报,2006,21(1):6-11. 被引量：20
4Mandelbrot B B. A multifractal walk down Wall Street[ J]. Scientific American, 1999, 298(1) : 70-73.
5Wei Y, Huang D-S, Multifractal analysis of SSEC in Chinese stock market : A different empirical result from Heng Seng index[J]. Physica A, 2005, 355(5):497-508.
6Plerou V, Gopikrishnan P, Amaral A, et al. Scaling of the distribution of price fluctuations of individual companies [ J ]. Phys. Rev. E, 1999, 60(3) : 6519-6529.
7Gopikrishnan P, Plerou V, Amaral A, et al. Scaling of the distributions of fluctuations of financial market indices [ J ]. Phys. Rev. E, 1999, 60(3): 5305-5316.
8Schmitt F, Schertzer D, Lovejoy S. Multifractal fluctuations in finance [ J ]. Int. J. Theor. Appl. Fin., 2000, 3 ( 3 ) : 361-364.
9Bacry E, Delour J, Muzy F. Modeling financial time series using muhifractal random walks [ J ]. Physica A, 2001, 299 (5) : 84-92.
10Sun X, Chen H, Wu Z, et al. Multifractal analysis of Hang Seng index in Hong Kong stock market[ J ]. Physica A, 2001, 291(7) : 553-562.