摘要
在全球化市场逐渐成形的大氛围下,不同国家金融系统之间的关联关系越来越强,我国期货市场日益成熟,上海期货交易所(SHFE)的期铜市场也成为了全球铜市需求方的定价中心,其与伦敦金属交易所(LME)之间的关联性也越来越强。全球性金融危机的爆发使得学术界、业界、监管层等各方市场参与者对期铜市场风险波动的重视程度加大,也给研究危机状况下上海期货交易所与伦敦金属交易所两市期铜市场间的风险溢出效应提供了一个契机。
本文在上述背景以及前人研究成果的基础上,首先借助EGARCH模型估算2000年至2009年整十年间LME与SHFE两市期铜三种常用置信水平下的多空头市场风险价值(VaR),并对两市VaR值相关性做进一步分析。然后以两市VaR值为着力点研究分析两市风险溢出效应。研究数据证实两市期铜市场各置信水平下的多空头VaR值之间存在很强的关联性,即LME与SHFE两市期铜市场之间风险相关性的存在得到肯定。因此论文结合样本期整十年间经济形势的起伏波动不同状况,分三个子样本期实证分析两市期铜市场间风险溢出效应。主要包括运用Granger因果关系检验法定性检验描述两市多空头市场风险溢出的Granger关系,检验LME期铜市场与SHFE期铜市场之间的风险溢出方向;然后根据不同样本期内两市多空头VaR值数据序列平稳性和协整性的不同特点,从无约束向量自回归模型或有约束向量自回归模型出发对不同样本期内各置信水平下的两市期铜风险溢出过程做简单定量的脉冲响应分析。
研究结果表明,整个研究样本期间内,LME与SHFE两市期铜多空头市场互为风险Granger原因,但在全球性金融危机发生前后,LME期铜多空头市场风险则单向是SHFE期铜多空头市场风险Granger原因;脉冲响应分析显示,金融危机发生前后LME与SHFE期铜市场面对彼此的风险溢出主要都呈正向递减冲击反应,其中还出现其它样本区间都没有的负向冲击反应,整个研究样本期内,LME与SHFE期铜市场间风险溢出则主要呈正向递增反应。而且研究结果还显示我国汇率制度的改革并未对LME与SHFE两市期铜市场之间风险溢出关系的整体性质造成影响。
本文主要研究创新点表现在以2000年初到2009年底整十年间最新的LME与SHFE两市期铜交易数据为研究数据支撑;在整十年的研究期内,本研究面临中国汇率制度根本性改革以及全球性金融危机发生这些前人所没有遇到的复杂研究环境;相比前人大都从价格以及收益率数据出发研究LME与SHFE两市期铜市场关联性,本文从两市期铜VaR值出发研究两市期铜市场风险关系。
根据研究思路,论文主体框架大致如下:
第一章为前言导论部分。首先介绍论文研究的背景与选题意义,再对国内外有关研究文献做一个简单综述并介绍论文主要研究思路和结构安排,指出论文的主要创新之处和所用研究方法。
第二章则主要介绍期铜市场风险度量方法和研究数据的选择,并对两市数据做初步处理。还简单说明了影响期铜市场的微观和宏观方面的供求因素。
第三章对LME与SHFE期铜市场关联度做检验,主要是对以EGARCH模型为基础估算出的两市VaR值之间的关联性做检验,包括GARCH类模型的建立检验选择、两市VaR值的估算和两市风险值之间关联性的检验。
第四章主要内容为LME与SHFE两市期铜风险溢出效应实证分析。包括用Granger因果关系检验法和脉冲响应分析法对两市风险溢出的方向及溢出过程影响的分析。
第五章则主要对论文研究内容做一个简单总结,并进一步指出论文研究不足之处,指出后续可能的研究方向。
Because of development of Chinese futures market, the correlation between Shanghai Futures Exchange's (SHFE) and London Metal Exchange (LME) becomes more and more strong. As a result, copper future market in SHFE has become the demand pricing center of the world copper market. At the same time, the global financial crisis leads market participants, such as academia, industry, and regulators, to pay more attention on risk fluctuation and gives opportunity to study risk spillover effect of copper futures between LME and SHFE.
Basing on the mentioned background above and previous literatures, this paper uses EGARCH model to estimate the LME and SHFE copper futures' value at risk (VaR) with their price data from 2000 to 2009, and the whole period is divided into three sub-sample stages according to different economic conditions. The paper analyzes further correlation and risk spillover effects between LME and SHFE with VaR calculated. It shows that copper futures of LME and SHFE have strong risk-related correlation in short and long position. And the paper analyzes the risk spillover effects between LME and SHFE in the whole sample period and three sub-sample stages. Before the analysis of risk spillover effects, we use granger causality model to test risk spillover granger relationship between the copper futures markets of LME and SHFE by vector autoregression model (VAR) and vector error correction model (VEC). In this paper, we also use impulse response function to test the two copper future markets' risk spillover volatility in each sample stage.
The results indicate that there exists significant two-way risk spillover granger causality between the two copper future markets in the whole period. But around the period of financial crisis, the two markets only have one-way risk spillover granger causality from LME to SHFE significantly. Impulse response analysis results show that VaR spillover is positive increasing and there are some short negative spillover effects which is never appeared in other sample stage. Furthermore, we know from these findings that Chinese exchange regime reform have no impacts on the risk spillover relationship between two copper future markets.
The innovations in this paper include: Using the latest transaction data of two copper future markets of LME and SHFE from 2000 to 2009; During the whole decade, our study faces exchange regime reform and global financial crisis which former scholars have never encountered; Comparing with former studies on relationship between two copper future markets with price and return, we use VaR series of two copper future markets.
Main framework of this paper is as follows: Chapter I is introduction. Chapter II mainly describes data selecting and processing. Chapter III uses GARCH model to estimate the copper future VaR in order to test risk relationship between two markets. Chapter IV is the empirical chapter to analyze risk spillover effects with Granger causality test and impulse response function. Chapter V summarizes the most contents briefly and points out several possible ongoing research directions.
引文
①潘慧声.金融危机冲击期铜牛市终结[N].期货日报,2008-10-10(6).
①刘向丽,成思危,汪寿阳,洪永淼.期现货市场间信息溢出效应研究[J].管理科学学报,2008,11(3):125-139.
①陈雄兵,张宗成.再议Granger因果检验[J].数量经济技术经济研究,2008(1):154-161.
②特伦斯·C·米尔斯.金融时间序列的计量经济学模型[M].北京:经济科学出版社,2002:9-57,218-320.
[1]李艺,汪寿阳.大宗商品国际定价权研究[M].北京:科学出版社,2007:58-82.
[2]特伦斯·C·米尔斯.金融时间序列的计量经济学模型[M].北京:经济科学出版社,2002:9-57, 218-320.
[3]辛乔立,孙兆东.次贷危机[M].北京:中国经济出版社,2008:15-52.
[4]陈雄兵、张宗成.再议Granger因果检验[J].数量经济技术经济研究,2008(1):154-161.
[5]方毅,张屹山.国内外金属期货市场“风险传染”的实证研究[J].金融研究,2007,5: 133-146.
[6]高辉.中国上海与英国伦敦商品期货价格的协整分析[J].哈尔滨师范大学自然科学学报,2004,20(2):22-26.
[7]华仁海,陈百助.国内、国际期货市场期货价格之间的关联研究[J].经济学(季刊),2004,3(3):727-742.
[8]华仁海,陈百助.涨跌停板制度对期货市场价格发现过程及波动性的影响——基于上海期货交易所的实证研究[J].数量经济技术经济研究,2006,5:86-93.
[9]华仁海、刘庆富.国内外期货市场之间的波动溢出效应研究[J].世界经济,2007,6:64-74.
[10]华仁海,仲伟俊.对我国期货期货市场价格发现功能的实证分析[J].南开管理评论,2002,5:57-61.
[11]洪永淼,成思危,刘艳辉,汪寿阳.中国股市与世界其他股市之间的大风险溢出效应[J].经济学(季刊),2004,3(3):705-728.
[12]蒋祥林,阳桦,吴晓霖.上海期货交易所铜期货价格发现功能研究——基于体制转换时间序列模型[C].2008年度上海市社会科学界第六届学术年会文集(经济·管理学科卷),上海:上海人民出版社,2008:654-661.
[13]陆凤彬,李艺,王栓红,汪寿阳.中国上海燃料油期货市场信息溢出研究——基于CCF方法与ECM模型[J].系统工程理论与系统,2008,3:25-44.
[14]刘海龙,黄伟.上海期铜定价的实证检验[J].管理科学学报,2007,10(3):52-57.
[15]李海英,马卫锋,罗婷.上海燃料油期货价格发现功能研究——基于GS模型的实证分析[J].财贸研究,2007,2:104-108.
[16]刘庆富,王海民.期货市场与现货市场之间的价格研究——中国农产品市场的经验[J].财经问题研究,2006,4:44-51.
[17]刘庆富,张金清,华仁海.LME与SHFE金属期货市场之间的信息传递效应研究[J].管理工程学报,2008,22(2):155-159.
[18]刘向丽,成思危,汪寿阳,洪永淼.期现货市场间信息溢出效应研究[J].管理科学学报,2008,11(3):125-139.
[19]李勇,李传乐.风险传染效应在牛市、熊市中的异化现象——来自A+H双重上市公司的证据[J].金融发展研究,2008,11:62-65.
[20]毛承雄,高翔.灰参数Verhulst模型在电力期货价格预测中的应用[J].水电能源科学,2005,23(3):80-82.
[21]马超群,佘升翔,陈彦玲,王振全.中国上海燃料油期货市场信息溢出研究[J].管理科学学报,2009,12(3):92-101.
[22]马正兵.我国粮食期货价格发现功能的交叉谱实证研究[J].统计与决策,2005,10:64-66.
[23]潘慧峰,张金水.国内外石油市场的极端风险溢出检验[J].中国管理科学,2007,15(3):25-30.
[24]宋军,李鹏.期铜市场的国际跨市套利的实证研究[J].中国金融评论,2007,1(4):25-41.
[25]史晓姗,吴振翔.我国铜期货市场与伦敦铜期货市场间信息传递的GIR-GARCH模型及分析[R].北京:长盛基金管理有限公司,2007.
[26]邵延平.GARCH模型对期铜市场风险的研究[J].运筹与管理,2007,16(2):108-112.
[27]王家辉.上海期货交易所金属铜国际定价能力研究[J].上海金融,2008,6:48-51.
[28]徐剑刚.期货报酬时间序列统计特性[J].统计研究,1997,3:70-72.
[29]吴文锋,刘太阳,叶中行.上海与伦敦期铜市场风险变异性实证研究[J].系统工程方法应用,2006,15(3):256-259.
[30]徐信忠,杨云红,朱彤.上海期货交易所铜期货价格发现功能研究[J].财经问题研究,2005,10:23-31.
[31]严太华,刘昱洋.我国商品期货价格与现贷价格协整关系的实证研究[J].预测,1999,18(3):72-74.
[32]张家豪,刘建和.中英期货市场铜期货跨市套利行为研究[J].商业经济,2010,3:1-2,86.
[33]赵进文.我国期货市场与国际期货市场关联度分析与协整检验[J].中国软科学,2004,5:34-40.
[34]夏天,程细玉.国内外期货价格与国产现货价格动态关系的研究[J].金融研究,2006,2:110-117.
[35]周颖,仇晓光.基于CVaR-GARCH-GED模型的单品种期货风险价值预测[J].统计与决策,2007,9:95-97.
[36]潘慧声.金融危机冲击期铜牛市终结[N].期货日报,2008-10-10(6).
[37]贺晋兵.石油期货市场风险问题研究——基于燃料油期货市场的统计分析[D].厦门:厦门大学,2008.
[38]蒋翠侠.动态金融风险测度及管理研究[D].天津:天津大学,2007.
[39]苏佳瑜.上证A股、上证B股和港股市场的动态相关性和金融危机传染性实证分析[D].厦门:厦门大学,2009.
[40]郑葵方.国际铜期货信息传递与资产定价[D].北京:北京航空航天大学博士论文,2007.
[41]郑秀田.我国黄金市场的波动特征与风险度量研究[D].杭州:浙江工商大学,2008.
[42] Baillie R.T.,G.G.Booth,Y.Tse,T.Zabotina.Price discovery and common factor models[J].Journal of Financial Markets,2002,5:309–321.
[43] Bekaert G.,C.R.Harvey.Emerging equity market volatility [J] .Journal of Financial Economics,1997,43 (1) :29-77.
[44] Cheung Y.,L .K.Ng .A causality-in-variance test and its application to financial market prices[J].Journal of Econometrics,1996,72:33-48.
[45]Engle R.F.,W.J.Granger . Co-integration and error correction: representation, estimation, and testing[J].Econometrica,1987,1(55):251-276.
[46] Hansen B.E..Autoregressive conditional density estimation[J].International Economic Review,1994,35(3):705-730.
[47] Hashmi A.R.,A.S.Tay.Global regional sources of risk in equity markets: evidence from factor models with time-varying conditional skewness[J].Journal of International Money and Finance,2007,26(3):430-453.
[48] Harvey C.R.,A.Siddique.Autoregressive conditional skewness[J].Journal of Financial and Quantitative Analysis,1999,34(4):465-487.
[49] Hong Y.M..A Test for Volatility Spillover with Applications to Exchange Rates[J].Journal of Econometrics,2001,103 ,183—224.
[50] Hong Y.M..Granger Causality in Risk and Detection of Risk Transmission Betwee- n Financial Markets.Working paper,Department of Economics and Department of Statistical Science,Cornell University ,2001.
[51] Hong Y.M. Detecting extreme risk spillover between financial markets[R].Working Paper of Cornell University,2003.
[52] Jondeau E.,M.Rockinger.Conditional volatility,skewness,and kurtosis: existence, persistence,and comovements[J].Journal of Economic Dynamics & Control,2003,27:1699-1737.
[53] Kawaller I.,P.Koch,T.Koch.The Temporal Price Relationship between S&P500 Futures and the S&P500 Index [J].Journal of Finance,1987,42:1309-1329.
[54] Korkie B.,R.Sivakumar,H.J.Turtle.Variance spillover and skewness in financial asset returns[J].The Financial Review,2006,41(1):139-156.
[55] Leon A.,G.Rubio,G.Serna .Autoregresive conditional volatility,skewness and kurtosis[J].The Quarterly Review of Economics and Finance,2005,45:599-618.
[56] Roope M.,R.Zurbruegg.The Intraday Price Discovery Process between the Singapore Exchange and Taiwan Futures Exchange [J] . Journal of Futures Markets,2002,22:219-240.
[57] Stoll H.R.,R.Whaley.The Dynamics of Stock Index and Stock Index Futures Returns[J].Journal of Financial and Quantitative Analysis,1990,25:441-468.
[58] Garbade K.D.,W.L.Silber.Price movements and price discovery in futures and cash markets[J].RevJew of Economi cs an d Statistics,1983,1(65):289-297.