期货市场日内VaR测度模型与应用
|
摘要
本文构建了两类日内VaR测度模型,一类是以超高频数据为基础,结合久期模型、波动模型和Monte Carlo模拟方法的综合日内VaR(IVaR)测度模型,另一类是以等时间间隔高频数据为基础并结合传统计量方法(历史模拟法与GARCH法)的日内VaR测度模型。然后运用上海燃料油期货市场数据进行了实证研究,结果表明:相对于传统计量方法,IVaR模型由于包含了更充分的市场信息,因而无论是多头头寸还是空头头寸时都具有更好的预测能力;IVaR方法估计的VaR值最小,说明IVaR模型比较适用于风险承受能力较强的投资者;IVaR模型对于空头头寸的管理更加严格;另外,IVaR模型的预测结果表明市场在日内具有开盘大,随后迅速衰减并趋于稳定的特征。
This paper built two kinds of measure model of intraday VaR,one was based on ultra-high frequency data,combined duration model,volatility model and Monte Carlo simulation method to build a comprehensive intraday VaR(IVaR) measure model,the other was based on the high frequency data of fix time interval,and on the traditional methods(Historical simulation method and GARCH method).Then this paper made the empirical study by using Shanghai fuel oil futures market data.The results indicate that,because the ultra-high frequency data includes more market information,either long or short position,IVaR model has better predictive ability relative to the traditional methods with fix time interval.The VaR value estimated by IVaR method is minimum,this shows that IVaR model is especially applicable to the investors with stronger risk tolerance.IVaR model is stricter with the management when in short position.In addition,IVaR has the intraday feature that IVaR value is larger at opening time,and will decrease rapidly and stabilize.
This paper built two kinds of measure model of intraday VaR,one was based on ultra-high frequency data,combined duration model,volatility model and Monte Carlo simulation method to build a comprehensive intraday VaR(IVaR) measure model,the other was based on the high frequency data of fix time interval,and on the traditional methods(Historical simulation method and GARCH method).Then this paper made the empirical study by using Shanghai fuel oil futures market data.The results indicate that,because the ultra-high frequency data includes more market information,either long or short position,IVaR model has better predictive ability relative to the traditional methods with fix time interval.The VaR value estimated by IVaR method is minimum,this shows that IVaR model is especially applicable to the investors with stronger risk tolerance.IVaR model is stricter with the management when in short position.In addition,IVaR has the intraday feature that IVaR value is larger at opening time,and will decrease rapidly and stabilize.
引文
[1]Engle R F,Russel J R.Autoregressive conditional duration:A new model for irregularly spaced transaction data[J].Econometrica,1998,66(5):1127-1162.
[2]Giot P.Market risk models for intraday data[J].European Journal of Finance,2005,(11):309-324.
[3]Pacurar Dionne G,Duchesne P,Maria.Intraday value at risk(IVAR)using tick-by-tick data with application to the Toronto stock exchange[J].Journal of Empirical Finance,2009,16(5):777-792.
[4]Engle R F.The econometrics of ultra-high frequency data[J].Econometrica,2000,68(1):1-22.
[5]Bauwens L,Giot P.The logarithmic ACD model:An application to the bid-ask quote process of three NYSE stocks[J].Annals of Economics and Statistics,2000,60(4):117-149.
[6]Ghysels E,Jasiak J.GARCH for irregurally spaced financial data:The ACD-GARCH model[J].Studies in Nonlinear Economics and Econometrics,1998,2:133-149.
[7]刘伟,陈敏,吴武清.高频数据交易量久期与价格变化的动态行为研究[J].数理统计与管理,2010,29(3):518-528.
[8]朱慧明,曾惠芳,郝立亚.基于MCMC的贝叶斯变结构金融时序GARCH模型研究[J].数理统计与管理,2011,30(6):1019-1017.
[9]Watsham Terry J.Futures and Options in Risk Management[M].北京:北京大学出版社,2003.
[10]王春峰.金融市场风险管理[M].天津:天津大学出版社,2000.
[11]Kupiec P.Techniques for verifyjng the accuracy of risk measurement models[J].Journal of Derivatives,1995,3(2):173-184.
[12]邵锡栋,连玉君,黄性芳.交易问隔、超高频波动率与VAR[J].统计研究,2009,26(1):96-102.
[2]Giot P.Market risk models for intraday data[J].European Journal of Finance,2005,(11):309-324.
[3]Pacurar Dionne G,Duchesne P,Maria.Intraday value at risk(IVAR)using tick-by-tick data with application to the Toronto stock exchange[J].Journal of Empirical Finance,2009,16(5):777-792.
[4]Engle R F.The econometrics of ultra-high frequency data[J].Econometrica,2000,68(1):1-22.
[5]Bauwens L,Giot P.The logarithmic ACD model:An application to the bid-ask quote process of three NYSE stocks[J].Annals of Economics and Statistics,2000,60(4):117-149.
[6]Ghysels E,Jasiak J.GARCH for irregurally spaced financial data:The ACD-GARCH model[J].Studies in Nonlinear Economics and Econometrics,1998,2:133-149.
[7]刘伟,陈敏,吴武清.高频数据交易量久期与价格变化的动态行为研究[J].数理统计与管理,2010,29(3):518-528.
[8]朱慧明,曾惠芳,郝立亚.基于MCMC的贝叶斯变结构金融时序GARCH模型研究[J].数理统计与管理,2011,30(6):1019-1017.
[9]Watsham Terry J.Futures and Options in Risk Management[M].北京:北京大学出版社,2003.
[10]王春峰.金融市场风险管理[M].天津:天津大学出版社,2000.
[11]Kupiec P.Techniques for verifyjng the accuracy of risk measurement models[J].Journal of Derivatives,1995,3(2):173-184.
[12]邵锡栋,连玉君,黄性芳.交易问隔、超高频波动率与VAR[J].统计研究,2009,26(1):96-102.