基于改进的GARCH模型对VaR风险研究
|
摘要
当前,GARCH模型普遍地被应用在金融资产序列波动性的预测,以及在险价值VaR的计算及市场风险管理中。针对股票市场的特点,采用将交易量与价格极差加入传统的GARCH模型中的方法对上证综合指数进行研究,发现不仅可以改善部分GARCH模型的拟合和预测结果,而且对于金融市场中标的资产的GARCH效应的解释能力逐渐降低,甚至有些资产标的物GARCH效应直接消失了,其中交易量和价格极差作为重要代理变量在关于收益率波动持续性方面表现出良好的解释作用,同时计算向前一步预测的在险价值VaR并对计算出的结果进行检验。实证研究表明,改进后的GARCH模型预测的VaR值相比于传统的GARCH模型计算结果更加准确,降低了VaR失效的概率,使得预测得到的VaR值与实际结果更加接近。
Currently,the GARCH model is widely used in the prediction of the volatility of financial asset sequences and the calculation of VaR at risk and market risk management.According to the characteristics of the stock market,using the method of adding the trading volume and price range difference to the traditional GARCH model to study the Shanghai Composite Index,we found that not only the fitting and forecasting results of some GARCH models can be improved,but also the assets of the underlying financial market.The GARCH effect's explanatory ability gradually decreases and even some asset objects GARCH effect disappears directly.Among them,the trading volume and price difference show a good explanation for the persistence of the yield volatility.At the same time,the VaR value of the forward one step prediction is calculated and calculated.The results are backtested.The results show that the VaR value predicted by the improved GARCH model is more accurate than the traditional GARCH model,reducing the probability of VaR failure and making the estimated VaR closer to the actual results.
Currently,the GARCH model is widely used in the prediction of the volatility of financial asset sequences and the calculation of VaR at risk and market risk management.According to the characteristics of the stock market,using the method of adding the trading volume and price range difference to the traditional GARCH model to study the Shanghai Composite Index,we found that not only the fitting and forecasting results of some GARCH models can be improved,but also the assets of the underlying financial market.The GARCH effect's explanatory ability gradually decreases and even some asset objects GARCH effect disappears directly.Among them,the trading volume and price difference show a good explanation for the persistence of the yield volatility.At the same time,the VaR value of the forward one step prediction is calculated and calculated.The results are backtested.The results show that the VaR value predicted by the improved GARCH model is more accurate than the traditional GARCH model,reducing the probability of VaR failure and making the estimated VaR closer to the actual results.
引文
[1]Rusy S.Tsay.金融时间序列分析[M].北京:人民邮电出版社,2012:95-121.
[2]Robert F.Engle.Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation[J].Economrteica,1982,(4):987-1007.
[3]Garman M.B.,Klass M.J.On the estimation of security price volatilities from historical data[J].Journal of Business,1980,(53):61-65.
[4]Parkinson M.The extreme value method for estimating the variance of rate of return[J].Journal of Business,1980,(53):61-65.
[5]Clark P.K.A subordinated stochastic process model with finite variances for speculative prices[J].Economrteica,1973,(41):135-136.
[6]Tauchen G.,Pitts M.The price variability-volume relationship on speculative markets[J].Economrteica,1983,(51):485-505.
[7]Gallo G.,Pacini.The effect of trading activity on market volatility[J].The European Journal of Finance,2000,(6):163-175.
[8]Alizadeh S.,Brandt M W,and Diebold F X.Range-based estimation of stochastic volatility models[J].Journal of Finance,2002,(57):1047-1092.
[9]Timotheos Angelidisa,Alexandros Benosa.The use of GARCH models in VaR estimation[J].Statistical Methodology,2004,(1):105-108.
[10]Mehmet Orhana,Bulent Koksal.A comparison of GARCH models for VaR estimation[J].Expert Systems with Applications,2012,(39):3582-3592.
[11]孙便霞,王明进.基于价格极差的GARCH模型[J].数理统计与管理,2013,(2):259-267.
[12]郑文通.金融风险管理的VaR法及其应用[J].国际金融研究,1999,(9):58-62.
[13]杨继平,袁璐,等.基于结构转换非参数GARCH模型的VaR估计[J].管理科学学报,2014,(17):69-80.
[14]杨炘,王邦宜.交易量与股价波动性:对中国市场的实证研究[J].系统工程学报,2005,(5):530-534.
[15]Kupiec H.P.Techniques for verifying the accuracy of risk measurement models[J].Journal of Derivatives,1995,(3):73-84.
[16]Peng H.,Lu Z.Nonlinear analysis of finance systems:Exploring the nonlinear impact of the trading volume on the price volatility[J].Journal of Systems Science and Mathematical Science,2009,(11):1527-1541.
[2]Robert F.Engle.Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation[J].Economrteica,1982,(4):987-1007.
[3]Garman M.B.,Klass M.J.On the estimation of security price volatilities from historical data[J].Journal of Business,1980,(53):61-65.
[4]Parkinson M.The extreme value method for estimating the variance of rate of return[J].Journal of Business,1980,(53):61-65.
[5]Clark P.K.A subordinated stochastic process model with finite variances for speculative prices[J].Economrteica,1973,(41):135-136.
[6]Tauchen G.,Pitts M.The price variability-volume relationship on speculative markets[J].Economrteica,1983,(51):485-505.
[7]Gallo G.,Pacini.The effect of trading activity on market volatility[J].The European Journal of Finance,2000,(6):163-175.
[8]Alizadeh S.,Brandt M W,and Diebold F X.Range-based estimation of stochastic volatility models[J].Journal of Finance,2002,(57):1047-1092.
[9]Timotheos Angelidisa,Alexandros Benosa.The use of GARCH models in VaR estimation[J].Statistical Methodology,2004,(1):105-108.
[10]Mehmet Orhana,Bulent Koksal.A comparison of GARCH models for VaR estimation[J].Expert Systems with Applications,2012,(39):3582-3592.
[11]孙便霞,王明进.基于价格极差的GARCH模型[J].数理统计与管理,2013,(2):259-267.
[12]郑文通.金融风险管理的VaR法及其应用[J].国际金融研究,1999,(9):58-62.
[13]杨继平,袁璐,等.基于结构转换非参数GARCH模型的VaR估计[J].管理科学学报,2014,(17):69-80.
[14]杨炘,王邦宜.交易量与股价波动性:对中国市场的实证研究[J].系统工程学报,2005,(5):530-534.
[15]Kupiec H.P.Techniques for verifying the accuracy of risk measurement models[J].Journal of Derivatives,1995,(3):73-84.
[16]Peng H.,Lu Z.Nonlinear analysis of finance systems:Exploring the nonlinear impact of the trading volume on the price volatility[J].Journal of Systems Science and Mathematical Science,2009,(11):1527-1541.