摘要
多元化投资导致不同行业及金融市场间的交互影响越来越大。对于投资者而言,理解不同行业指数之间的相互关系极为重要,投资者可根据冲击传导的方向和影响的大小去判断其他行业的价格变化,进而进行有效的风险预防和投资管理。因此,市场参与者对行业之间存在怎样的波动影响也越来越感兴趣。
波动溢出效应是指一个市场的波动不仅受到自身前期波动的影响,而且受到其他市场历史波动的影响,这种波动在市场之间的传递称为市场间的波动溢出效应。从风险投资和管理角度看,行业间股价的溢出效应会给投资和管理带来额外的风险。为了分散、化解、转移这种联动的风险,常常需要涉及对多个不同行业间股价的溢出效应进行研究,从而实现资产的组合、风险的对冲、市场的管理等。因而,考察沪市不同行业的股价波动的溢出效应,对于研究我国证券市场的结构、投资组合、风险度量和管理、资产配置、政策监管都具有重要的意义。
这篇文章选取了中国股市制造、地产、金融、石化四个行业指数的日收益序列作为研究对象,将它们两两分为六组,分别进行建立多元GARCH模型,进行波动溢出分析。其时间区间为：2001年7月3日至2010年7月16日共计2190个数据。
论文的第一部分是绪论,简单的介绍了选题背景以及波动溢出和多元GARCH模型的国内外研究综述,同时介绍了单变量GARCH类模型的发展。在第二部分介绍了一些检验的理论知识并详细阐述了多元GARCH模型。论文的第三部分是实证部分,首先介绍衡量波动性的各种指标以及数据来源,之后利用这些指标衡量各个行业指数的统计特征,随后采用ADF和PP检验方法检验各个收益率序列的平稳性,并将各个序列应用于自回归异方差模型(ARCH)以确定其波动特征。对每个行业指数收益率序列建模之后,又进行了二元GARCH建模。将四个行业收益率序列两两分为六组,分别建立了均值与方差方程。逐个对模型估计结果进行分析,分析行业间的波动溢出效应。最后把四个收益率序列合在一起建立四元GARCH模型。论文的第四部分阐述了行业间波动溢出效应的实际经济意义,并相应的提出了一些建议。有助于读者对不同行业间冲击传导有初步的认识,有利于进行合理的资产组合,降低风险。
通过对各个行业收益率序列的建模,得出以下结论：四个行业的波动都显著的受到自身前期值波动的影响,波动具有聚类性和较强的持续性。且各个行业之间存在显著的波动溢出效应。
本文的创新之处是：已有的研究很少触及同一证券市场中不同行业的股票收益率之间波动溢出关系,运用多元GARCH模型进行研究的也是较少。本文试图通过多元GARCH模型对中国股市四个支柱行业收益与波动的关系的研究。对不同行业的股票之间的收益和波动关系的研究有助于实现有效地金融投资资产分布和建立行业收益模型。其研究结果可以为关心股市波动在行业间如何扩散并进行投资组合的普遍投资者和银行提供有益的参考。
本文的不足之处是,侧重从实证的角度出发进行研究,对行业之间的波动溢出经济理论分析不够深入；另外,对行业间波动溢出的分析还有待进一步的深入研究,可以用多种模型从不同的角度进行分析。但是限于文章篇幅和时间的原因,本文没有涉及。
Diversification of investment has lead to more interaction in different sectors and between financial markets increasingly. It's extremely important for investors to understand the relationships of different sectors. Investors can judge the size of changes in other sectors according to the direction of the impact and influence. And then carry out effective risk prevention and investment management. So financial markets participants are more and more interested in knowing how shocks and volatility are transmitted across markets over time.
Volatility spillover effect is that the volatility of a market not only be influenced by their early fluctuations, but also by the historical volatility of other markets. From the perspective of venture capital and management, inter-industry spillovers price will bring additional investment and risk management. In order to disperse, dissolve, transfer the risk of such a linkage is often necessary to research the spillover effect between the different sectors. Such can achieve a combination of assets, risk hedging, the market management. So, researching the characteristics of different sectors volatility spillover effects is important for the study of the structure of China's securities market, portfolio, risk measurement and management, asset allocation, policy regulation.
This paper selects the Chinese stock market's manufacturing, the real estate, finance, petrochemical four industry index return series for the study data, they divided into six groups with two and two in one group, establishing a diversity of GARCH model for volatility spillover analysis. The time interval:July 3,2001 to July 16,2010, a total of 2190 data.
In the first part is an introduction, a simple introduction to the topics, background and volatility spillover GARCH Model Research at home and abroad, Then introduces the univariate GARCH model development. In the second part describes some of the test methods theoretical and the multivariate GARCH model in detail. The third part is the empirical part, at first introduced the sources of data and indicators which measure of volatility, and then use these indicators to measure the statistical characteristics of each industry index, Then using the ADF and PP test methods to test the sequence of all the stationary rate of return, and each sequence used in ARCH model to determine the wave characteristics. Then for each industry index return series, the GARCH model has been made. Every model estimation results were analyzed. In the fourth part discusses the inter-industry volatility spillover effects of the real economic significance, and accordingly put up some proposals. This will help the readers have a preliminary understanding of the impact of transmission between the different sectors and reducing the risk.
Through model to every sector return series, we can get the following conclusions:the volatility of the four industries is subject to significant fluctuations in the value of their own early, with the clustering of volatility and strong sustainability. Between the different sectors have significant volatility spillover effects.
Innovation of this paper is:very little research has been hit the same relationship between the different sectors of the stock return, not to say using the M-GARCH model. This paper attempts to research four pillar industries'spillover relationships in China's stock market returns, to fill existing gaps in the documentary. Researching on the volatility relationships between different sectors of the stock return can help to achieve effective distribution of financial assets and the establishment of industry revenue model. Its findings can provide a useful reference on how the stock market volatility spread between the industry and the general investment portfolio investors and banks.
Shortcomings of this article are that focusing on the perspective from the empirical study of the industry Volatility Spillover. Then the theoretical analysis was not enough; In addition, the analysis of inter-industry volatility spillover remains to be further in-depth research, we can use a variety of models (such as DCC model) from different angles to research. With the limitation in space and time, this article does not contain these.
引文
[1]陈浪南,童汉飞,洪如明.《波动率研究》[M].第一版.北京：中国财政经济出版社，2008.
[2]董秀良,曹凤岐.《国内外股市波动溢出效应-基于多元GARCH模型的实证研究》[J].数理统计与管理,2009(6)：1091-1099.
[3]董秀良,吴仁水.《交易量适合作为股价波动信息的代理变量吗?——来自中国沪深股市的证据》[J].数量经济技术经济研究,2008(1)：97.108.
[4]樊智、张世英.《多元GARCH建模及其在中国股市分析中的应用》[J].管理科学学报,2003(2)：68-76.
[5]范致镇,陈秀权.《汇率与股价的溢出效应及长期联动性一一基于汇改后中国市场的实证研究》[J].证券市场,2010(1)：91-95.
[6]高铁梅.《计量经济分析方法与建模：Eviews应用及实例》[M].第二版.北京：清华大学出版社,2006.
[7]李昆.《上海证券交易所行业指数的收益扩散和波动扩散效应》[J].经济体制改革,2003,(02).
[8]李文君,尹康.《多元GARCH模型研究述评》[J].2009,(10)：138-146.
[9]王彬,李成,马文涛.《国际石油价格与通货膨胀的溢出效应及动态相关性》[J].财经研究,2010(4)：25-35.
[10]王明进.《多元波动率模型的一些新进展》[J].数理统计与管理,2010,(3)：232-247.
[11]王群勇,王国忠.沪市A、B股市场间信息传递模式研究[J].现代财经,2005(6)：26-29.
[12]Ruey S.Tsay.《金融时间序列分析》[M].北京：机械工业出版社,2006.
[13]吴奉刚,王芙蓉.《中国股市与汇市的波动溢出效应研究》[J].金融教学与研究,2008(6)：52-55.
[14]吴相俊,陈讯.《我国股市与汇市波动溢出动态性实证研究》[J].经济前沿,2009(12)：28-34.
[15]肖喻,晓庆宪.《多元GARCH模型在国内企业债券波动传递研究中的应用》[J].上海理工大学学报,2007(5)：485-490.
[16]姚奕,曹广喜.《沪深股市动态溢出效应与动态相关性的实证研究——基于长记忆VAR-BEKK(DCC)-MVGARCH(1,1)模型》[J].系统工程2008(5)：47-54.
[17]殷玲,唐杰.《GARCH-M模型与我国沪深股市的波动》[J].江南大学学报,2002(2)：56-58.
[18]张剑,王一鸣,等.涨跌停板制度对我国股市影响的实证研究(2002)[J].经济科学,2002(3)：34-48.
[19]张瑞锋.金融市场协同波动溢出分析及实证研究[J].数量经济技术研究,2006(10)：141-149.
[20]张世英,樊智.《协整理论与波动模型—金融时间序列分析及应用》[M].北京：清华大学出版社,2004.
[21]张晓峒《EViews使用指南与案例》[M].机械工业出版社,2006.
[22]周少甫,杜福林.《基于多元GARCH理论的资本资产定价模型》[J].决策参考,2004(12)：47-48.
[23]赵华《人民币汇率与利率之间的价格和波动溢出效应研究》[J].金融研究,2007(3)：41-49.
[24]Bolleslev T. Generalized autoregressive conditional heteroskedasticity [J] Journal of Econometrics,1986,31:307-327.
[25]Bollerslev T. Modelling the coherence in short-run nominal exchange rates:A multivariate generalized ARCH approach[J].Review of Economics and Statistics,1990(72):498-505.
[26]Bollerslev T, Engle R F, Wooldridge J M. A capital asset pricing model with time varying covariance[J]. The Journal of Political Economy(96):116-131.
[27]Engle R F. Autoregressive conditional heteroskedasticity with estimate of the variances of UK inflation[J]. Econometrica,1982(31):987-1008.
[28]Engle R. Estimating time varying risk premia in the term structure:The ARCH-M model[J]. Econometrica,1987,55:391-407.
[29]Engle R. F. Dynamic conditional correlation:a simple class of multivariate Generalized autoregressive conditional heteroskedasticity models[J]. Journal of Business and Economic Statistics,2002(20):339-350.
[30]Engle R F, Kroner K F. Multivariate simultaneous generalized ARCH[J]. Econometric Theory,1995(11):122-150.
[31]Harris, Richard D.F. Stock Markets and Development:A Reassessment. European Economic Review,1997,(1):156-163.
[32]Hassan SAa, Malik F.2007.Multivariate GARCH modeling of sector volatility transmission[J]. The Quarterly Review of Economics and Finance,47(3),470-480.
[33]Hong Li, Ewa Majerowska. Testing stock market linkages for Poland and Hungary:A multivariate GARCH approach[J]. Research in International Business and Finance,2008(22):247-266.
[34]Kevin James Daly. Financial Volatility and Real Economic Activity [M]. Ashgate Publishing Ltd,1999.
[35]King M, and Wadhwani S. Transmission of volatility between stock markets[J].The Review of Financial Studies,1990,3(1):5-33.
[36]Robert F. Engle. Dynamic Conditional Correlation:A Simple Class of Multivariate Generalized Autoregressive Conditional Heteroskedasticity Models,2002,20(3):339-350.
[37]Suhejla Hoti, Michael McAleer, Laurent L. Pauwels. Multivariate volatility in environmental finance[J]. Mathematics and Computer in Simulation,2008(78).
[38]Tse Y. K. and A. K. Tsui. A Multivariate Generalized Autoregressive Conditional Heteroscedasticity Model with Time-varying Correlations[J]. Journal of Business and Economic Statistics,2002(20):351-362.
[39]Y. K. Tse and A. K. Tsui, A multivariate garch model with time-varying correlations[J]. National University of Singapore, December 1998.