基于高频数据的中国股票市场流动性度量研究
摘要
流动性是度量市场效率最基本的核心指标,是股票市场的基本属性。流动性的测算研究和影响流动性的因素研究,对投资者和市场监管者都意义重大,这是本文选题的依据和出发点。
本文首先将买卖价差、交易量和结合了价量关系的流动性比率等指标进行对比分析,对其分布特征和相关性进行了实证研究。实证结果表明中国股票市场股票流动性度量指标不服从常见的分布函数;流动性的指标的相关性实证研究表明有充分的证据否定所有的指标之间相关性高的结论。特别是对于价格类指标和交易量类指标而言,这两者非常可能出现对流动性度量得出不同甚至相反的结论。
然后本文对基于交易量持续期的流动性度量进行了研究。本文在研究过程中,对整体和买方和卖方三类交易量持续期进行了研究,发现这三类交易量持续期都存在倒“W”型的日内模式,并对日内模式进行剔除,然后采用EACD、WACD和LOG-ACD模型对三类交易量持续期分布进行估计,最后对估计模型进行拟合检验,最后结果揭示:无论EACD、WACD、还是LOG-ACD模型都较好的拟合了高频数据的持续性特征,能够反映交易的聚集性和日内特征。WACD模型,对于高频数据的交易时间间隔序列拟合得最好。EACD模型在预测交易时间间隔、瞬时交易频率方面比较方便,因为该模型可以写为线性的ARMA形式,所以EACD(1,1)模型在高频数据的建模研究中是使用也较多的。而Log-ACD模型则在预测发展趋势上有着很好的效果,对于长期流动性有着很好的预测。
Liquidity is one of the most basic factors to measure the performance of the market and is the basic properties of the stock market. Research and is great significance for investors and market regulators, which is the just reason for this article topics.
Firstly, this article will use bid-ask spread, depth indicators and indicators combining the price with the volume to analysis their distribution and relevance. Empirical results show that China's stock market indicators for the stock liquidity measure do not obey common distribution function. The research shows that there is ample evidence of the indicators to deny the high correlation between indicators. Especially for indicators of the price and trading volume, both them are very likely to present different result for the same situation.
Then, based on the duration of trading, liquidity measurement is studied. In this paper, we discussed the duration of trading of the overall, the buyers and the sellers, and found that the "W"-shaped pattern of days exist in the three kinds of duration, at the same time, we eliminated the pattern. Afterwards, we used EACD, WACD and LOG-ACD model to estimate the duration of the three trading volume, and at last accomplished the testing of fitness. The final result revealed that: no matter EACD, WACD, or LOG-ACD model are preferably fitted the duration character of high-frequent data, and can reflect the accumulation of transaction and the character of in-day. Thereunto, WACD model is best to fit the transaction time interval of high-frequency data; while EACD is easier to predict the trading time interval and instantaneous transaction frequency, because the model can be written as a linear form of ARMA, the EACD (1,1) model in the modeling of high-frequency data is used commonly. Log-ACD model is good at forecasting the developing trend and has good results for long-term mobility.
本文首先将买卖价差、交易量和结合了价量关系的流动性比率等指标进行对比分析,对其分布特征和相关性进行了实证研究。实证结果表明中国股票市场股票流动性度量指标不服从常见的分布函数;流动性的指标的相关性实证研究表明有充分的证据否定所有的指标之间相关性高的结论。特别是对于价格类指标和交易量类指标而言,这两者非常可能出现对流动性度量得出不同甚至相反的结论。
然后本文对基于交易量持续期的流动性度量进行了研究。本文在研究过程中,对整体和买方和卖方三类交易量持续期进行了研究,发现这三类交易量持续期都存在倒“W”型的日内模式,并对日内模式进行剔除,然后采用EACD、WACD和LOG-ACD模型对三类交易量持续期分布进行估计,最后对估计模型进行拟合检验,最后结果揭示:无论EACD、WACD、还是LOG-ACD模型都较好的拟合了高频数据的持续性特征,能够反映交易的聚集性和日内特征。WACD模型,对于高频数据的交易时间间隔序列拟合得最好。EACD模型在预测交易时间间隔、瞬时交易频率方面比较方便,因为该模型可以写为线性的ARMA形式,所以EACD(1,1)模型在高频数据的建模研究中是使用也较多的。而Log-ACD模型则在预测发展趋势上有着很好的效果,对于长期流动性有着很好的预测。
Liquidity is one of the most basic factors to measure the performance of the market and is the basic properties of the stock market. Research and is great significance for investors and market regulators, which is the just reason for this article topics.
Firstly, this article will use bid-ask spread, depth indicators and indicators combining the price with the volume to analysis their distribution and relevance. Empirical results show that China's stock market indicators for the stock liquidity measure do not obey common distribution function. The research shows that there is ample evidence of the indicators to deny the high correlation between indicators. Especially for indicators of the price and trading volume, both them are very likely to present different result for the same situation.
Then, based on the duration of trading, liquidity measurement is studied. In this paper, we discussed the duration of trading of the overall, the buyers and the sellers, and found that the "W"-shaped pattern of days exist in the three kinds of duration, at the same time, we eliminated the pattern. Afterwards, we used EACD, WACD and LOG-ACD model to estimate the duration of the three trading volume, and at last accomplished the testing of fitness. The final result revealed that: no matter EACD, WACD, or LOG-ACD model are preferably fitted the duration character of high-frequent data, and can reflect the accumulation of transaction and the character of in-day. Thereunto, WACD model is best to fit the transaction time interval of high-frequency data; while EACD is easier to predict the trading time interval and instantaneous transaction frequency, because the model can be written as a linear form of ARMA, the EACD (1,1) model in the modeling of high-frequency data is used commonly. Log-ACD model is good at forecasting the developing trend and has good results for long-term mobility.
引文
[1] Acharya V.V and Pedersen L.H, Asset Pricing with Liquidity Risk[R].NBER Working Paper, 2002
[2] Admati A.R, and Pfleiderer P., A theory of intraday patterns: volume and price variability[J]. Review of Financial Studies,1988,1(1):3-40
[3] Bacidore, Jeffrey M., Robert H. Battalio, and Robert H. Jennings, Depth Improvement and Adjusted Price Improvement on the New York Stock Exchange[R].NYSE Working Paper,2000
[4] Bagehot WA, The only game in town [J].Financial Analysis Journal, 1971,27(2):12~14
[5]Balestra P., Nerlove M.Pooling Cross Section and Time Series Data in theEstimation of a Dynamic Model: TheDemand for Natural Gas [J]. Econometrica, 1966, 34(3):585~612
[6]Barclay M.J., Smith C.W., Corporate Payout Policy: Cash Dividends Versus Open-Market Share Repurchases [J].Journal of Financial Economics, 1988, 22(1): 61~82.
[7]Bessembinder H., Trade execution costs on Nasdaq and the NYSE: A post-reform comparison [J]. Journal of Financial and Quantitative Analysis, 1999,34: 387~408.
[8]Black F. Toward a Fully Automated Stock Exchange: Part I[J]. Financial Analysis Journal, 1971, 27: 28-35
[9]Breen W.J, Hodrick L.S., Korajczyk R.A., Predicting Equity Liquidity[R].Working paper, Kellogg Graduate School of Management, Northwestern University, 2000
[10]Brennan M.J., and Subrahmanyam A., Market microstructure and asset pricing:on the compensation for illiquidity in stock returns[J]. Journal of Financial Economics, 1996, 41: 441~464
[11]Engle, Robert F and Jeffery R.Russell,1998, Autoregressive Conditional Duration: A New Model for Irregularly Spaced Transaction Data,Econometrica 66:1127-1162.
[12]Engle,Robert ,The Econometrics of Ultra High Frequency Data, Econometrica 2000, 68:1-22.
[13]Ghysels, Eric and Joanna Jasiak, 1998, GARCH for Irregularly Spaced Data:the ACD-GARCH Model,Studies in Nonlinear Dynamics and Econometrics, 133.149
[14]Ghysels.E,Gourieroux.C and Jasiak.J,1998,Stochastic Volatility Duration Models,working paper,Olsen and Associates High Frequency Data in FinanceⅡ,conference Proceedings.
[15]Zhang, Michael Y, Jeffrey R.Russell and Ruey S.Tsay, 2001,A Nonlinear Autoregressive Conditional Duration Model with Applications to Financial Transaction Data,Journal of Econometrics 104:179-207.
[16]Engle, Robert and Asger Lunde,1998,Trades and quotes:A bivariate point process,Discussion Paper,University of California,San Diego.
[17]Bauwens, Veredas, D., 1999, The stochastic conditional duration model:A latent factor model for the analysis of nancial durations,forthcoming in Journal of Econometrics.
[18]Engle, Robert F.and Russell,J.R.,1997,Forecasting the frequency of changes in quoted foreign exchange prices with the autoregressive conditional duration model,Journal of Empirical Finance 4,187-212.
[19]Hasbrouck, Joel, 1991, Measuring the information content of stock trades,Journal of Finance 46:179-207
[20]Dufour, Alfonso and Robert F.Engle, 2000, Time and the price impact of trade,Journal of Finance 55:2467-2498
[21]Robert F.Engle.and Andrew J., 2000, Patton, Impacts of trades in an error-correction model of quote prices, University of California at San Diego, Economics Working Paper Series: 2000-26
[22]Ghysels,Eric and Joanna Jasiak,1998,GARCH for Irregularly SPaced Financial Data:The ACD一GARCH Model,Working PaPer,CIRANO
[23]Bauwens,L.,Giot,P.,2000,The logarithmic ACD model:An application to the bid-ask quote Process of three NYSE stocks,Annalesd’Economie et de Statistique,60,117一150.
[24]Engle, Robert F.and Jeffery R.Russell, 2002, Analysis of High Frequency Data, Working Paper, New York University and University of Chicago
[25]Bufenglin and Zhangweiguo,Empirical study on time effect and price impact of trades in Shanghai stock market,重庆大学学报(英文版)2004.03 132-135
[26]Silvia Golia,Long memory effects in ultra-high frequency data,Quaderni di Statistica,2001,3:43-52
[27]房振明,王春峰,曹媛媛,上海证券市场流动性模式的研究[J],管理工程学报,2005, 2:33~39
[28]韩冬,中国证券市场流动性风险研究[M],社会科学文献出版社,2007,05:18-23
[29]靳云汇,杨文,上海股市流动性影响因素实证分析[J],金融研究,2002,06:12~21
[30]梁朝晖,张维,流动性的期权定价方法[J],北京航空航天大学学报(社会科学版),2005b,18(3):8~11
[31]梁朝晖,张维,证券流动性折扣的期权定价方法——封闭式基金折价的流动性分析[J],西南交通大学学报(社会科学版),2005a,6(1):84~87
[32]粱朝晖,张维,任达,基于最优清算策略的流动性风险溢价测算[J],北京理工大学学报(社会科学版),2006,8(2):61~63
[33]刘狄《如何衡量流动性:理论与文献综述》,《上证研究》vol.1,2:115.137
[34]孙培源.施东晖.微观结构、流动性与买卖价差:一个基于上海股市的经验研究世界经济2002.04:69-72
[35]蒋学雷,陈敏,王国明,吴国富.股票市场的流动性度量的动态ACD模型[J].统计研究,2004,(4) :42一44
[36]马超群,张明良.中国证券市场的LOG_ ACD模型及其应用[J].统计与决策,2006 ,2(下):92-95
[37]补冯林、张卫国、何伟基于超高频数据的股票流动性度量研究[J].统计与决策2005.02:24-27
[38]施东晖,上海证券市场日内流动性与波动性研究,中国证券报,2001,8:10-14
[39]赵留彦,王一鸣,沪深股市交易量与收益率及其波动的相关性:来自实证分析的证据[J].经济科学,2003,2:57-67
[40]刘海龙,吴冲锋,金融市场微观结构理论综述[J].金融研究,2003,15(4):41-44 [41J约翰.Y.坎贝尔,安德鲁.w.罗.金融市场计量经济学.上海:上海财经大学出版社,2003,48一66
[42]李汉东,张世英.波动持续性对资产定价模型的影响分析[J].亚太金融研究,2002,(4):19一23
[43]石洪华,郑伟.上海证券市场日期效应检验及其有效性的研究[J].系统工程理论与实践,2003,(5):41一45
[44]马超群,张明良.中国证券市场的LOG-ACD模型及其应用研究[J].统计与决策, 2006,(2):92-95
[45]陈敏,王国明等.中国证券市场的ACD-GARCH模型及其应用[J].统计研究,2003(12):34-39
[46]陈怡玲,宋逢明.中国股市价格变动与交易量关系的实证研究[J].管理科学学报,2000(3):56-61
[47]许睿,刘海龙,吴冲锋,2004.中国A股市场流动性统计特性与变化趋势,系统工程理论与实践,2004(3):38-42
[48]庄新田,刘洋,中国股票市场流动性实证研究[J],南方经济,2006,2:71~79
[49]闫东鹏,吴贵生,流动性和条件资产定价:理论和实证检验[J].南方经济,2006,6:63~74
[50]杨朝军,孙培源,施东晖,微观结构、市场深度与非对称信息:对上海股市日内流动性模式的一个解释[J].世界经济,2002,11:53-58
[51]鲁万波,ACD模型及其扩展—金融高频数据计量模型的新动态[J].统计研究,2005(10):7-9
[52]张晓峒,EVIEWS使用指南与案例[M],北京,机械工业出版社,2007
[53]张晓峒,计量经济分析[M],天津,南开大学出版社,2005
[2] Admati A.R, and Pfleiderer P., A theory of intraday patterns: volume and price variability[J]. Review of Financial Studies,1988,1(1):3-40
[3] Bacidore, Jeffrey M., Robert H. Battalio, and Robert H. Jennings, Depth Improvement and Adjusted Price Improvement on the New York Stock Exchange[R].NYSE Working Paper,2000
[4] Bagehot WA, The only game in town [J].Financial Analysis Journal, 1971,27(2):12~14
[5]Balestra P., Nerlove M.Pooling Cross Section and Time Series Data in theEstimation of a Dynamic Model: TheDemand for Natural Gas [J]. Econometrica, 1966, 34(3):585~612
[6]Barclay M.J., Smith C.W., Corporate Payout Policy: Cash Dividends Versus Open-Market Share Repurchases [J].Journal of Financial Economics, 1988, 22(1): 61~82.
[7]Bessembinder H., Trade execution costs on Nasdaq and the NYSE: A post-reform comparison [J]. Journal of Financial and Quantitative Analysis, 1999,34: 387~408.
[8]Black F. Toward a Fully Automated Stock Exchange: Part I[J]. Financial Analysis Journal, 1971, 27: 28-35
[9]Breen W.J, Hodrick L.S., Korajczyk R.A., Predicting Equity Liquidity[R].Working paper, Kellogg Graduate School of Management, Northwestern University, 2000
[10]Brennan M.J., and Subrahmanyam A., Market microstructure and asset pricing:on the compensation for illiquidity in stock returns[J]. Journal of Financial Economics, 1996, 41: 441~464
[11]Engle, Robert F and Jeffery R.Russell,1998, Autoregressive Conditional Duration: A New Model for Irregularly Spaced Transaction Data,Econometrica 66:1127-1162.
[12]Engle,Robert ,The Econometrics of Ultra High Frequency Data, Econometrica 2000, 68:1-22.
[13]Ghysels, Eric and Joanna Jasiak, 1998, GARCH for Irregularly Spaced Data:the ACD-GARCH Model,Studies in Nonlinear Dynamics and Econometrics, 133.149
[14]Ghysels.E,Gourieroux.C and Jasiak.J,1998,Stochastic Volatility Duration Models,working paper,Olsen and Associates High Frequency Data in FinanceⅡ,conference Proceedings.
[15]Zhang, Michael Y, Jeffrey R.Russell and Ruey S.Tsay, 2001,A Nonlinear Autoregressive Conditional Duration Model with Applications to Financial Transaction Data,Journal of Econometrics 104:179-207.
[16]Engle, Robert and Asger Lunde,1998,Trades and quotes:A bivariate point process,Discussion Paper,University of California,San Diego.
[17]Bauwens, Veredas, D., 1999, The stochastic conditional duration model:A latent factor model for the analysis of nancial durations,forthcoming in Journal of Econometrics.
[18]Engle, Robert F.and Russell,J.R.,1997,Forecasting the frequency of changes in quoted foreign exchange prices with the autoregressive conditional duration model,Journal of Empirical Finance 4,187-212.
[19]Hasbrouck, Joel, 1991, Measuring the information content of stock trades,Journal of Finance 46:179-207
[20]Dufour, Alfonso and Robert F.Engle, 2000, Time and the price impact of trade,Journal of Finance 55:2467-2498
[21]Robert F.Engle.and Andrew J., 2000, Patton, Impacts of trades in an error-correction model of quote prices, University of California at San Diego, Economics Working Paper Series: 2000-26
[22]Ghysels,Eric and Joanna Jasiak,1998,GARCH for Irregularly SPaced Financial Data:The ACD一GARCH Model,Working PaPer,CIRANO
[23]Bauwens,L.,Giot,P.,2000,The logarithmic ACD model:An application to the bid-ask quote Process of three NYSE stocks,Annalesd’Economie et de Statistique,60,117一150.
[24]Engle, Robert F.and Jeffery R.Russell, 2002, Analysis of High Frequency Data, Working Paper, New York University and University of Chicago
[25]Bufenglin and Zhangweiguo,Empirical study on time effect and price impact of trades in Shanghai stock market,重庆大学学报(英文版)2004.03 132-135
[26]Silvia Golia,Long memory effects in ultra-high frequency data,Quaderni di Statistica,2001,3:43-52
[27]房振明,王春峰,曹媛媛,上海证券市场流动性模式的研究[J],管理工程学报,2005, 2:33~39
[28]韩冬,中国证券市场流动性风险研究[M],社会科学文献出版社,2007,05:18-23
[29]靳云汇,杨文,上海股市流动性影响因素实证分析[J],金融研究,2002,06:12~21
[30]梁朝晖,张维,流动性的期权定价方法[J],北京航空航天大学学报(社会科学版),2005b,18(3):8~11
[31]梁朝晖,张维,证券流动性折扣的期权定价方法——封闭式基金折价的流动性分析[J],西南交通大学学报(社会科学版),2005a,6(1):84~87
[32]粱朝晖,张维,任达,基于最优清算策略的流动性风险溢价测算[J],北京理工大学学报(社会科学版),2006,8(2):61~63
[33]刘狄《如何衡量流动性:理论与文献综述》,《上证研究》vol.1,2:115.137
[34]孙培源.施东晖.微观结构、流动性与买卖价差:一个基于上海股市的经验研究世界经济2002.04:69-72
[35]蒋学雷,陈敏,王国明,吴国富.股票市场的流动性度量的动态ACD模型[J].统计研究,2004,(4) :42一44
[36]马超群,张明良.中国证券市场的LOG_ ACD模型及其应用[J].统计与决策,2006 ,2(下):92-95
[37]补冯林、张卫国、何伟基于超高频数据的股票流动性度量研究[J].统计与决策2005.02:24-27
[38]施东晖,上海证券市场日内流动性与波动性研究,中国证券报,2001,8:10-14
[39]赵留彦,王一鸣,沪深股市交易量与收益率及其波动的相关性:来自实证分析的证据[J].经济科学,2003,2:57-67
[40]刘海龙,吴冲锋,金融市场微观结构理论综述[J].金融研究,2003,15(4):41-44 [41J约翰.Y.坎贝尔,安德鲁.w.罗.金融市场计量经济学.上海:上海财经大学出版社,2003,48一66
[42]李汉东,张世英.波动持续性对资产定价模型的影响分析[J].亚太金融研究,2002,(4):19一23
[43]石洪华,郑伟.上海证券市场日期效应检验及其有效性的研究[J].系统工程理论与实践,2003,(5):41一45
[44]马超群,张明良.中国证券市场的LOG-ACD模型及其应用研究[J].统计与决策, 2006,(2):92-95
[45]陈敏,王国明等.中国证券市场的ACD-GARCH模型及其应用[J].统计研究,2003(12):34-39
[46]陈怡玲,宋逢明.中国股市价格变动与交易量关系的实证研究[J].管理科学学报,2000(3):56-61
[47]许睿,刘海龙,吴冲锋,2004.中国A股市场流动性统计特性与变化趋势,系统工程理论与实践,2004(3):38-42
[48]庄新田,刘洋,中国股票市场流动性实证研究[J],南方经济,2006,2:71~79
[49]闫东鹏,吴贵生,流动性和条件资产定价:理论和实证检验[J].南方经济,2006,6:63~74
[50]杨朝军,孙培源,施东晖,微观结构、市场深度与非对称信息:对上海股市日内流动性模式的一个解释[J].世界经济,2002,11:53-58
[51]鲁万波,ACD模型及其扩展—金融高频数据计量模型的新动态[J].统计研究,2005(10):7-9
[52]张晓峒,EVIEWS使用指南与案例[M],北京,机械工业出版社,2007
[53]张晓峒,计量经济分析[M],天津,南开大学出版社,2005
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |