全球视野的大类资产风险溢出研究被引量：9

Research on the Risk Spillover of Large Class Assets in the Global Perspective

导出

摘要随着金融国际化程度不断加深,金融风险跨市场、跨区域的传染不断加剧,如何识别风险、追溯风险源是管理风险和投资决策的重要前提。利用修正的已实现波动率建立VAR模型,以广义误差方差分解矩阵构建时变风险溢出指数和风险溢出网络,从全球视野研究大类资产之间的风险溢出行为。实证研究2000年以来外汇市场、股票市场和期货市场的27种重要资产,通过静态分析和滚动分析对大类资产的风险溢出展开详实分析。研究结果表明,①通过对8种汇率的研究发现,外汇市场存在很强的风险溢出效应。欧美汇率是外汇市场最主要的风险传染源,其余汇率长期承受外部市场的风险传染;人民币汇率受外部市场冲击较小,2014年以来其溢出指数逐渐上升。②通过对11个股市的研究发现,全球股市之间的联动性极强。美国股市一直是全球股市最重要的风险传染源,全球股市长期承受欧美股市的风险冲击;中国股市2015年的暴跌对全球股市造成较大的风险冲击。③通过对6种商品期货的研究发现,伦敦金属期货联动性较强,原油期货风险溢出较强,而贵金属期货在牛熊市中产生持续性的风险传染。④通过对27种资产交叉溢出的研究发现,全球股票市场、外汇市场和期货市场之间具有很强的风险溢出关系。欧美汇率和美国股市对全球金融市场的风险溢出效应最强,期货品种的风险传染相对较弱。研究结果兼具理论意义和实践意义。在理论方面,风险溢出指数和风险溢出网络为金融资产的风险溢出行为提供了强大的研究工具。在实践方面,投资者应关注中外股市风险传染,也要充分重视汇率波动风险。在政策方面,为制定汇率政策和防范股市风险提供了经验证据。为投资策略、资产配置和风险管理研究提供有益参考。 With the deepening of financial internationalization,the financial riskspillover across different markets or regions has intensified.How to identify risks and trace risk sources is an important prerequisite for managing risks and making investment decisions.The VAR model was established using modified realized volatility,and the time-varying spillover index and risk spillover network are constructed by generalized error variance decomposition matrix.The risk spillover of large class assets was studied from the global perspective.This study empirically studies 27 important assets in the foreign exchange market,stock market and futures market since 2000.Through static analysis and rolling analysis,the volatility spillover effects of large class assets are analyzed in detail.The main conclusions of this study:①Research on eight exchange rates shows that there is a strong volatility spillover effect in the foreign exchange market.The EUR/USD exchange rate is the most important risk source in the foreign exchange market,and the remaining exchange rates have long been exposed to the risks of other markets;the RMB exchange rate is less affected by other markets,and its spillover index has gradually increased since 2014.②A study of 11 stock markets shows that the volatility linkage between global stock markets is extremely strong.The US stock market has always been the most important risk source in the global stock markets.Quite a few stock markets have long suffered from the risk shocks of American or European stock markets.The plunge of the Chinese stock market in 2015 has caused a greater risk impact to the global stock markets.③Research on six commodity futures demonstrates that London metal futures have strong linkages,crude oil futures exhibit strong volatility spillovers,gold and silver futures have generated sustained risk contagion in the large bull market.④Examining the 27 assets,it is found that there is a strong volatility spillover effects between global stock markets,foreign exchange markets and futures markets.EUR/USD exchange rate and the S&P500 index have the strongest spillover effects on the global financial markets,and the risk transmission of futures products is relatively weak.This study has both theoretical and practical significance.In theory,the time-varying spillover indexes and risk spillover network provide powerful research tools for investigating volatility spillovers of financial assets.In terms of investment,investors should not only pay attention to the risk impact fromthe oversea stock markets,but also the risk of exchange rate markets.In terms of policy,this study provides empirical evidence for the formulation of exchange rate policy,and has guiding significance for the prevention of stock market risks.Overall,this study provides a useful reference for investment strategy,asset allocation and risk management.

作者陈声利赵学军张自力 CHEN Shengli;ZHAO Xuejun;ZHANG Zili(Guanghua School of Management,Peking University,Beijing 100871,China;Harvest Fund Management Co.,Ltd.,Beijing 100005,China)

机构地区北京大学光华管理学院嘉实基金管理有限公司

出处《管理科学》 CSSCI 北大核心 2019年第6期3-17,共15页 Journal of Management Science

基金中国博士后科学基金一等资助项目(2018M640367)。

关键词大类资产已实现波动率溢出指数滚动分析风险溢出网络 large class assets realized volatility spillover index rolling analysis risk spillover network

分类号 F830.9 [经济管理—金融学]

作者简介陈声利,管理学博士,北京大学光华管理学院与嘉实基金管理有限公司联合培养的博士后研究员,研究方向为人工智能和量化投资等,代表性学术成果为“基于跳跃、好坏波动与百度指数的股指期货波动率预测”,发表在2018年第2期《系统工程理论与实践》,E-mail:chenshengli_hit@qq.com;赵学军,经济学博士,嘉实基金管理有限公司董事长,研究方向为行为金融等,代表性学术成果为“中国股市惯性策略和反转策略的实证分析”,发表在2001年第6期《经济研究》,E-mail:zhaoxuejun@jsfund.cn;张自力,物理学博士,嘉实基金管理有限公司董事总经理,研究方向为量化投资等,代表性学术成果为“股票网络、系统性风险与股票定价”,已被《经济学(季刊)》录用,E-mail:zhangzl@jsfund.cn。