摘要
尾部相依性刻画了金融资产在不同市场间发生大幅度同步涨跌的可能性。由于不同市场抗风险的能力及风险传染的路径不同,当已知一个市场出现极值现象时,考察其他市场是否也会出现极值现象,对深入理解不同市场间的相依结构有重要的意义。文章应用多元条件极值模型,考察了伦敦与上海市场期货铜在2008年金融危机前后的下尾部相依性,发现两个市场存在正的尾部相依关系,且在危机的不同演进阶段,尾部相依程度随之发生变化。特别地,文章构建的模型除了可以计算当某个市场发生极端变化时另一市场发生极端变化的概率,还可以预测该市场资产的预期收益率及其未来可能出现的极值范围,这些将有助于为极端风险下的投资决策及市场监管提供定量的参考依据。
The tail dependence describes the possibility of extreme events simultaneously occurring in different markets.In case of an extreme event in one market,similar crisis may or may not happen in other markets.Studying the extreme market behavior conditionally may help us better understand the tail structure between different markets.In this article,we build a multivariate conditional extreme value model based on Heffernan and Tawn(2004)'s method to examine the lower tail dependence of copper future price differences between London and Shanghai markets.The results show that the two markets are positively dependent in the extremes,and their extremal dependence become stronger if there is a bear market.We also obtain the conditional mean that helps to predict the possible distribution for future extreme points.
作者
秦学志
郭明
黄劲松
QIN Xuezhi;GUO Ming;HUANG Jinsong(College of Economics and Management,Dalian University of Technology,Dalian 116024,China)
出处
《大连理工大学学报(社会科学版)》
CSSCI
北大核心
2019年第4期20-30,共11页
Journal of Dalian University of Technology(Social Sciences)
基金
国家自然科学基金项目:“基于或有资本的银行极端金融风险的激励相容式分担方法研究”(71471026)
关键词
尾部相依性
多元条件极值模型
期货铜
Tail Dependence
Multivariate Extreme Value Theory
Copper Future Price