中文摘要：

条件概率分布常用来研究马尔科夫序列相依模型的构建,组合资产的相依结构受多方面的影响,资产之间的相互影响与时间上的记忆效应是组合资产两类主要的相依关系.结合条件概率的理论建立基于Copula函数相依关系模型,研究组合资产之间同期相依关系及时间上的短期相依关系,提出了模型参数的三阶段极大似然估计方法.

英文摘要：

Conditional probability distributions have been commonly used in modeling Markov chains.The dependence structures of portfolios of assets are affected by many things. The impact of assets each other and the memory effect with respect to time are two primary classes of dependence relation.This paper investigates the contemporaneous dependence of portfolios of assets and the temporal dependence of time by constituting the copula-based dependence models and combining the conditional probability theory.The three-stage pseudo maximum likelihood estimation of the parameters is proposed.