摘要
自回归条件持续时间(ACD)模型在金融经济中的作用日益明显,研究成果日益丰富。本文将Copula方法与ACD模型有机结合,运用Copula理论建立多元Copula-ACD模型,描述交易持续时间的时变条件相关关系,在一定程度上缓解了向量ACD模型参数难估计与非同步交易的问题。基于上证50指数中四只银行成分股的实证分析表明多元Copula-ACD模型能够较好捕捉中国证券市场上交易持续时间的聚集结构特征,给出交易持续时间的联合密度函数,估计各持续时间变量之间的自相关性与截面相关性,进一步描述和检验多元持续时间的溢出效应,为投资者提供决策参考。
There are a growing body of literature demonstrating that ACD model play an increasingly role in the field of the finance and economics.By combing Copula method and ACD model,using the theory of Copula to build multivariate Copula-ACD model,in this paper the condition correlation relationship of trading duration is described.This model can alleviate the difficulty of estimating parameters and nonsynchronous trading in vector ACD model at some degree.The empirical analysis based on the four index stocks from the SSE 50 index,multivariate Copula-ACD model proved to capture the clustering structure better in trading duration in Chinese stocks market and provides the joint density function for trading duration.The model is used to estimate the autocorrelation and cross-sectional correlation and further describe and test the spillover effect of multivariate durations to help decision making.
出处
《中国管理科学》
CSSCI
北大核心
2014年第S1期293-298,共6页
Chinese Journal of Management Science
基金
国家自然科学基金资助项目(71101118)
教育部新世纪优秀人才支持计划资助项目(NCET-13-0961)