论文部分内容阅读
在已有动态Copula模型基础上,提出可同时描述尾部相依性的非对称和长记忆特征的Copula模型.基于沪深股市数据,首次从尾部相依性的角度检验了沪深股市的长记忆效应.研究发现,沪深两市在重大利好或利空消息冲击时的相关性(即尾部相依性)都具有长记忆效应,极端事件对尾部相依性的影响比对未来收益和波动的影响更加持久.而且,样本外分析结果表明,相比已有Copula模型,具有长记忆性的Copula模型能更准确地预测未来1周至1年的市场间相关性.
Based on the existing dynamic Copula model, a copula model with asymmetric and long memory characteristics describing tail dependency is proposed. Based on the data of Shanghai and Shenzhen stock markets, the long memory effect of Shanghai and Shenzhen stock markets is tested for the first time from the perspective of tail dependence. The study finds that the correlation between the two markets in the event of major or negative news impact (ie tail dependence) has a long memory effect, and the impact of extreme events on tail dependence is more lasting than the impact on future earnings and volatility. The results of the out-of-sample analysis show that the Copula model with long memory can predict the correlation between markets in the next 1 week to 1 year more accurately than the existing Copula model.