中国证券市场波动性预测研究
|
摘要
波动性是股票市场最基本的特性。如果价格恒定不变,则将不会有买卖价差,最终会没有交易。但是,由于股票市场的稳定性和有效性决定了投资者效用最大化的实现,过高的波动性势必将对股票市场的健康运行产生不利影响。但是大量的国内波动性研究中,我们发现运用收益率来衡量波动性的研究数不胜数,但却缺少了运用极差衡量证券市场波动性的研究,且现有的模型拟合市场波动性预测的准确性也不尽人意,并有待提高,尤其是国内极少使用AV模型来衡量市场波动性。国外的一些研究表明了,运用AV模型来描述波动性较已有的一些模型更为准确,目前国内有关这方面的研究较少。
本文采用GARCH模型和基于极差的自回归波动率模型衡量市场波动性预测的准确性,并探究我国证券市场非对称性波动的情况。首先通过收益率和极差两个数据进行统计分析,分析我国上证市场的波动特征和现状。其次,分别运用GARCH模型和AV模型对收益率和极差两组数据进行建模拟合,分析中国上证市场波动性预测,得到适合中国上证市场波动性预测的指标和模型。最后,分别EGARCH模型和AV-α模型对收益率和极差两组数据进行建模拟合,分析中国上证市场的非对称性,验证中国上证市场是否存在非对称性。同时得出适合分析中国上证市场非对称性的波动率指标和模型。并得到以下结论：上证市场股票价格存在着明显的尖峰厚尾性,且是有偏的；极差对于衡量中国证券市场的波动性较传统的收益率指标更为准确；GARCH模型并没有AV类模型对于中国证券市场波动性预测的准确,这也同时验证了AV模型在中国证券市场的运用。并且,两个模型都验证了中国证券市场存在着较弱的波动预测性,即中国证券市场并非有效的,是弱势有效的。中国证券市场存在着不对称性,同时得到了负面消息(利空消息)比正面消息(利好消息)对中国证券市场冲击要大。
Stock market volatility is the most basic characteristics. If the price is constant, there would be Bid-ask spread, and eventually there would be no deal. However as the stock market stability and the validity decide to achieve maximizing the utility to investors, high volatility will set an adverse effect on the health of the operation the stock market. In a large number of domestic volatility studies, we found that numerous studies used the rate of return to measure volatility, but without using the range to measure volatility of the stock market. At the same time, the market volatility forecasting accuracy of existing models is unsatisfactory, and they needs to be improved, especially domestic studies rarely use AV model to measure the market volatility. Some study overseas showed that the AV model to describe the volatility in the use of the existing models is more accurate, and the domestic researches on these aspects are seldom.
This paper uses GARCH model and AV model based on the range to measure the prediction of market volatility, and explores the asymmetry in our stock market volatility. Firstly through the statistic analysis of the yields and range, this paper analyzes the characteristics and the status quo of the volatility of the market. Secondly, using GARCH model and AV model respectively based on the return and the range, this paper builds simulation and analysis of China's Shanghai market volatility forecast and gets the volatility index and the prediction model which is suitable for China's Shanghai market. Finally, using EGARCH model and AV-α model respectively based on the return and the range, this paper builds simulation and analysis of China's Shanghai market asymmetric, validates China's Shanghai market whether there is asymmetrical. And getting the following conclusions:this paper thinks that the Shanghai market share price has the obvious peak thick tail sex, and is biased; it was found that the range to measure the volatility of the Chinese securities market is more accurate than the traditional return index. Meanwhile, in the process of analysis this paper discovered in the volatility of the traditional forecasting GARCH model and no AV model for China securities market volatility of accurate prediction. Through the empirical results of the asymmetry, it also found the range is more accurate than the traditional index yields. And, in the two model, the empirical results found that China's securities market exists asymmetry, and got negative news (bad news) make more impact to the China securities than positive messages (good news).
本文采用GARCH模型和基于极差的自回归波动率模型衡量市场波动性预测的准确性,并探究我国证券市场非对称性波动的情况。首先通过收益率和极差两个数据进行统计分析,分析我国上证市场的波动特征和现状。其次,分别运用GARCH模型和AV模型对收益率和极差两组数据进行建模拟合,分析中国上证市场波动性预测,得到适合中国上证市场波动性预测的指标和模型。最后,分别EGARCH模型和AV-α模型对收益率和极差两组数据进行建模拟合,分析中国上证市场的非对称性,验证中国上证市场是否存在非对称性。同时得出适合分析中国上证市场非对称性的波动率指标和模型。并得到以下结论：上证市场股票价格存在着明显的尖峰厚尾性,且是有偏的；极差对于衡量中国证券市场的波动性较传统的收益率指标更为准确；GARCH模型并没有AV类模型对于中国证券市场波动性预测的准确,这也同时验证了AV模型在中国证券市场的运用。并且,两个模型都验证了中国证券市场存在着较弱的波动预测性,即中国证券市场并非有效的,是弱势有效的。中国证券市场存在着不对称性,同时得到了负面消息(利空消息)比正面消息(利好消息)对中国证券市场冲击要大。
Stock market volatility is the most basic characteristics. If the price is constant, there would be Bid-ask spread, and eventually there would be no deal. However as the stock market stability and the validity decide to achieve maximizing the utility to investors, high volatility will set an adverse effect on the health of the operation the stock market. In a large number of domestic volatility studies, we found that numerous studies used the rate of return to measure volatility, but without using the range to measure volatility of the stock market. At the same time, the market volatility forecasting accuracy of existing models is unsatisfactory, and they needs to be improved, especially domestic studies rarely use AV model to measure the market volatility. Some study overseas showed that the AV model to describe the volatility in the use of the existing models is more accurate, and the domestic researches on these aspects are seldom.
This paper uses GARCH model and AV model based on the range to measure the prediction of market volatility, and explores the asymmetry in our stock market volatility. Firstly through the statistic analysis of the yields and range, this paper analyzes the characteristics and the status quo of the volatility of the market. Secondly, using GARCH model and AV model respectively based on the return and the range, this paper builds simulation and analysis of China's Shanghai market volatility forecast and gets the volatility index and the prediction model which is suitable for China's Shanghai market. Finally, using EGARCH model and AV-α model respectively based on the return and the range, this paper builds simulation and analysis of China's Shanghai market asymmetric, validates China's Shanghai market whether there is asymmetrical. And getting the following conclusions:this paper thinks that the Shanghai market share price has the obvious peak thick tail sex, and is biased; it was found that the range to measure the volatility of the Chinese securities market is more accurate than the traditional return index. Meanwhile, in the process of analysis this paper discovered in the volatility of the traditional forecasting GARCH model and no AV model for China securities market volatility of accurate prediction. Through the empirical results of the asymmetry, it also found the range is more accurate than the traditional index yields. And, in the two model, the empirical results found that China's securities market exists asymmetry, and got negative news (bad news) make more impact to the China securities than positive messages (good news).
引文
[1]Engle, R.F.,1982. Autoregressive conditional heteroskedasticity with estimates of the variance of UK inflation. Econometrica 50,987-1008.
[2]Bollerslev, T.,1986. Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics 31,307-327.
[3]Bollerslev, T. A Conditionally Heteroskedastic Time Series Model for Speculative Prices and Rates of Return[J]. Review of Economics and Statistics,1987, (69),542-547.
[4]Engle, Robert F, David M. Liken, and Russell P.Robins. Estimating time varying risk premia in the term structure: The ARCH-M model. Econometrics,1987.(55):391-407.
[5]Akgllie,V.,1989, Conditional Heteroskedasticity in Time Series of Stock Return: Evidence and Forecasts, Journal of Business 62,55-80.
[6]Bemd & Klaus, 1996. Variances of security price returns based on high, low, and closing prices. Journal of Business 56,97-112.
[7]Tse, Y.K.,1992,Lead-lag Relationship Between Spot Index and Futures Price of the Nikkei Stock Average, Journal of Forecasting 14,553-563.
[8]Dimson,Peter A.,March. Proposals to Restructure Social Security. Journal of Economics Perspectives 10,2 (Summer 1990):67-68
[9]Timorthy. Robert,1997, "Intra-daily Information of Range-based Volatility for MEM-GARCH", Mathematics and Computers in Simulation, (79):2625-2632.
[10]Parkinson, M.,1980, "The extreme value method for estimating the variance of the rate of return", Journal of Business,53:61-65.
[11]Chou, R.,2005, "Forecasting financial volatilities with extreme values:the Conditional Auto Regressive Range (CARR) Model", Journal of Money Credit and Banking, 37 (3):561-582.
[12]Martens, M., and D. Dijk, 2007, "measuring volatility with the realized range", Journal of Econometrics,138:181-207.
[13]Hongquan Li, Yongmiao Hong, 2011, "Financial volatility forecasting with range-based autoregressive volatility model", Finance Research Letters, 8 (2011):69-76.
[14]张永东,毕秋香.上海股市波动性预测模型的实证比较[J].管理工程学报,2003,(02).
[15]刘凤芹,吴喜之.基于SV模型的深圳股市波动的预测[J].山西财经大学学报;2004(04).
[16]刘金全,刘志刚.我国经济增长率与条件波动性的双区制状态划分与相关性分析[J].经济研究,2010,(02).
[17]赵振全；薛丰慧.股票市场交易量与收益率动态影响关系的计量检验：国内与国际股票市场比较分析[J];世界经济;2005(11).
[18]庞素琳,徐建闽,黎荣舟.BP算法和对称ARCH类模型对股市波动性预测的实证比较[J].控制理论与运用,2006,(14).
[19]鲁万波.基于非参数GARCH模型的中国股市波动性预测[J].数理统计与管理,2006,(14).
[20]宋琴.汇率与股价：基于上市公司的实证分析[J].金融发展研究;2006(2).
[21]顾锋娟GARCH模型和SV模型的运用比较研究——以上证指数的波动性为例[J].浙江万里学院学报,2007,(2).
[22]马兴杰.考虑基差非对称性效应的期货波动性预测研究[J].西南交通大学学报,2008,(4).
[23]巩兰杰,张龙斌.一种考虑基差非对称影响的期货波动性预测模型研究——基于上海铜期货市场的实证分析[J];北京理工大学学报(社会科学版).2008(04).
[24]张新前,胡日东.基于CARR模型和GARCH模型的股市波动性预测研究[J].西安财经学院学报,2008,(02).
[25]陈德华,石建民.我国证券市场风险波动性预测：基于沪深300指数的比较研究[J].财政与金融,2009,(04).
[26]赵伟雄,崔海蓉,何建敏GARCH类模型波动率预测效果评价——沪铜期货为例[J].西安电子科技大学学报,2010,(04).
[27]孟卫东,宋丽伟,阳军.基于GARCH模型的沪深地产股波动性分析及预测[J].决策参考,2010,(04).
[28]杨二鹏,张德生,李文静,邵慧.基于非参数GARCH模型的沪深300股指波动性研究[J].基础科学学报,2010,(01).
[29]唐勇,张世英.高频数据的加权已实现极差波动及其实证分析[J],系统工程.2006(08):52-57.
[30]丁忠明,夏万军.中国股市波动的CARR模型分析[J],商业经济与管理.2005(12):45-51.
[31]周杰,刘三阳,邵锡栋.基于样本分位数的波动率估计：条件自回归拟极差模型,南开经济研究.2007(5)：133-143.
[32]葛新权.对回归预测模型的改进[J].数量经济技术经济研究,1999,(06).
[33]郑梅,苗佳.对于预测上海股指波动性的数量模型的比较[J].华北科技学院学报,2007,(03).
[34]王春峰,庄泓刚,房振明,卢涛.多重指标波动性模型在中国股市波动性估计和预测的运用——基于高频数据的研究[J].北京理工大学学报,2008,(05).
[35]王翔坤.对于预测上海股指波动性的数量模型的比较非参数GARCH模型、参数GARCH模型估计波动性对比[J].财经论坛,2010,(06).
[37]胡海鹏,方兆本.股市波动性预测模型改进研究[J].数理统计与管理,2004,(05).
[38]李隽.关于波动率及其预测方法的文献综述[J].山西青年管理干部学院学报,2009,(01).
[39]郑周.基于不同分布假设GARCH模型对上证指数波动性预测能力的比较分析[J].价值工程,2004,(03).
[40]鲁万波.基于非参数GARCH模型的中国股市波动性预测[J].数理统计与管理,2006,(04).
[41]房振明,王春峰,付臣余.基于高频数据的估值风险预测[J].财经论坛,2010,(02).
[42]张楠,孙德山.基于支持向量回归的股市波动性预测[J].金融视线,2010,(04).
[43]邱冬阳,王涛,许雄奇.金融市场随机波动：基于文献综述的视角[J].决策参考,2009,(04).
[44]刘雪峰.考虑基差非对称性效应的期货波动性预测模型研究[J].北京科技大学学报,2010,(09).
[45]张永东,毕秋香.上海股市波动性预测模型的实证比较[J].管理工程学报,2003,(02).
[46]黄冠钥,陈睿.我国股市波动性预测的非线性研究——来自GARCH族模型的对比发现[J].技术经济与管理研究,2007,(02).
[47]魏宁,黄登仕.中国股票市场价格波动特征及其可预测性研究[J].管理工程学报,2004,(04).
[48]王进,何凯欣.我国股市波动性分析——以上证指数为例[J].金融之窗,2010,(17).
[49]刘凤芹.上海股市波动的预测方式和模型[J].数学的实践与认识,2004,(10).
[50]张永东,毕秋香.上海股市波动性预测模型的实证比较[J].管理工程学报,2003,(02).
[51]Alizadeh, S., Brandt, M.W., Diebold, EX.,2002. Range-based estimation of stochastic volatility models. Journal of Finance 57,1047-1091.
[52]Garman, M.B., Klass, M.J., 1980. On the estimation of price volatility from historical data. Journal of Business 53,67-78.
[53]Beckers, S.,1983. Variances of security price returns based on high, low, and closing prices. Journal of Business 56,97-112.
[54]Hsieh, D.A.,1995. Nonlinear dynamics in financial markets: evidence and implications. Financial Analysts Journal 51,55-62.
[2]Bollerslev, T.,1986. Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics 31,307-327.
[3]Bollerslev, T. A Conditionally Heteroskedastic Time Series Model for Speculative Prices and Rates of Return[J]. Review of Economics and Statistics,1987, (69),542-547.
[4]Engle, Robert F, David M. Liken, and Russell P.Robins. Estimating time varying risk premia in the term structure: The ARCH-M model. Econometrics,1987.(55):391-407.
[5]Akgllie,V.,1989, Conditional Heteroskedasticity in Time Series of Stock Return: Evidence and Forecasts, Journal of Business 62,55-80.
[6]Bemd & Klaus, 1996. Variances of security price returns based on high, low, and closing prices. Journal of Business 56,97-112.
[7]Tse, Y.K.,1992,Lead-lag Relationship Between Spot Index and Futures Price of the Nikkei Stock Average, Journal of Forecasting 14,553-563.
[8]Dimson,Peter A.,March. Proposals to Restructure Social Security. Journal of Economics Perspectives 10,2 (Summer 1990):67-68
[9]Timorthy. Robert,1997, "Intra-daily Information of Range-based Volatility for MEM-GARCH", Mathematics and Computers in Simulation, (79):2625-2632.
[10]Parkinson, M.,1980, "The extreme value method for estimating the variance of the rate of return", Journal of Business,53:61-65.
[11]Chou, R.,2005, "Forecasting financial volatilities with extreme values:the Conditional Auto Regressive Range (CARR) Model", Journal of Money Credit and Banking, 37 (3):561-582.
[12]Martens, M., and D. Dijk, 2007, "measuring volatility with the realized range", Journal of Econometrics,138:181-207.
[13]Hongquan Li, Yongmiao Hong, 2011, "Financial volatility forecasting with range-based autoregressive volatility model", Finance Research Letters, 8 (2011):69-76.
[14]张永东,毕秋香.上海股市波动性预测模型的实证比较[J].管理工程学报,2003,(02).
[15]刘凤芹,吴喜之.基于SV模型的深圳股市波动的预测[J].山西财经大学学报;2004(04).
[16]刘金全,刘志刚.我国经济增长率与条件波动性的双区制状态划分与相关性分析[J].经济研究,2010,(02).
[17]赵振全；薛丰慧.股票市场交易量与收益率动态影响关系的计量检验：国内与国际股票市场比较分析[J];世界经济;2005(11).
[18]庞素琳,徐建闽,黎荣舟.BP算法和对称ARCH类模型对股市波动性预测的实证比较[J].控制理论与运用,2006,(14).
[19]鲁万波.基于非参数GARCH模型的中国股市波动性预测[J].数理统计与管理,2006,(14).
[20]宋琴.汇率与股价：基于上市公司的实证分析[J].金融发展研究;2006(2).
[21]顾锋娟GARCH模型和SV模型的运用比较研究——以上证指数的波动性为例[J].浙江万里学院学报,2007,(2).
[22]马兴杰.考虑基差非对称性效应的期货波动性预测研究[J].西南交通大学学报,2008,(4).
[23]巩兰杰,张龙斌.一种考虑基差非对称影响的期货波动性预测模型研究——基于上海铜期货市场的实证分析[J];北京理工大学学报(社会科学版).2008(04).
[24]张新前,胡日东.基于CARR模型和GARCH模型的股市波动性预测研究[J].西安财经学院学报,2008,(02).
[25]陈德华,石建民.我国证券市场风险波动性预测：基于沪深300指数的比较研究[J].财政与金融,2009,(04).
[26]赵伟雄,崔海蓉,何建敏GARCH类模型波动率预测效果评价——沪铜期货为例[J].西安电子科技大学学报,2010,(04).
[27]孟卫东,宋丽伟,阳军.基于GARCH模型的沪深地产股波动性分析及预测[J].决策参考,2010,(04).
[28]杨二鹏,张德生,李文静,邵慧.基于非参数GARCH模型的沪深300股指波动性研究[J].基础科学学报,2010,(01).
[29]唐勇,张世英.高频数据的加权已实现极差波动及其实证分析[J],系统工程.2006(08):52-57.
[30]丁忠明,夏万军.中国股市波动的CARR模型分析[J],商业经济与管理.2005(12):45-51.
[31]周杰,刘三阳,邵锡栋.基于样本分位数的波动率估计：条件自回归拟极差模型,南开经济研究.2007(5)：133-143.
[32]葛新权.对回归预测模型的改进[J].数量经济技术经济研究,1999,(06).
[33]郑梅,苗佳.对于预测上海股指波动性的数量模型的比较[J].华北科技学院学报,2007,(03).
[34]王春峰,庄泓刚,房振明,卢涛.多重指标波动性模型在中国股市波动性估计和预测的运用——基于高频数据的研究[J].北京理工大学学报,2008,(05).
[35]王翔坤.对于预测上海股指波动性的数量模型的比较非参数GARCH模型、参数GARCH模型估计波动性对比[J].财经论坛,2010,(06).
[37]胡海鹏,方兆本.股市波动性预测模型改进研究[J].数理统计与管理,2004,(05).
[38]李隽.关于波动率及其预测方法的文献综述[J].山西青年管理干部学院学报,2009,(01).
[39]郑周.基于不同分布假设GARCH模型对上证指数波动性预测能力的比较分析[J].价值工程,2004,(03).
[40]鲁万波.基于非参数GARCH模型的中国股市波动性预测[J].数理统计与管理,2006,(04).
[41]房振明,王春峰,付臣余.基于高频数据的估值风险预测[J].财经论坛,2010,(02).
[42]张楠,孙德山.基于支持向量回归的股市波动性预测[J].金融视线,2010,(04).
[43]邱冬阳,王涛,许雄奇.金融市场随机波动：基于文献综述的视角[J].决策参考,2009,(04).
[44]刘雪峰.考虑基差非对称性效应的期货波动性预测模型研究[J].北京科技大学学报,2010,(09).
[45]张永东,毕秋香.上海股市波动性预测模型的实证比较[J].管理工程学报,2003,(02).
[46]黄冠钥,陈睿.我国股市波动性预测的非线性研究——来自GARCH族模型的对比发现[J].技术经济与管理研究,2007,(02).
[47]魏宁,黄登仕.中国股票市场价格波动特征及其可预测性研究[J].管理工程学报,2004,(04).
[48]王进,何凯欣.我国股市波动性分析——以上证指数为例[J].金融之窗,2010,(17).
[49]刘凤芹.上海股市波动的预测方式和模型[J].数学的实践与认识,2004,(10).
[50]张永东,毕秋香.上海股市波动性预测模型的实证比较[J].管理工程学报,2003,(02).
[51]Alizadeh, S., Brandt, M.W., Diebold, EX.,2002. Range-based estimation of stochastic volatility models. Journal of Finance 57,1047-1091.
[52]Garman, M.B., Klass, M.J., 1980. On the estimation of price volatility from historical data. Journal of Business 53,67-78.
[53]Beckers, S.,1983. Variances of security price returns based on high, low, and closing prices. Journal of Business 56,97-112.
[54]Hsieh, D.A.,1995. Nonlinear dynamics in financial markets: evidence and implications. Financial Analysts Journal 51,55-62.