论文部分内容阅读
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,资产之间的同期相依与单个资产时间上的短期相依是组合资产两类主要的相依关系.结合条件概率的理论,考虑组合资产之间的同期相依与时间上的短期相依两类关系,建立基于Copula函数相依关系模型研究了沪深股市指数收益率的相依结构.应用三阶段极大似然估计方法对模型的参数进行估计,应用χ~2检验统计量对模型进行优度检验和模型的比较.研究结果表明:考虑了单个资产时间上短期相依关系的模型更适合描述沪深股市的相依结构.
Conditional probabilistic distributions are often used to study the construction of Markov sequence-dependent models. The dependency structure of portfolio assets is influenced by many factors. The contemporaneous dependence between assets and the short-term dependence of individual assets on time are the two major types of portfolio asset dependencies. Combining with the theory of conditional probability, we consider the relationship between contemporaneous dependency and short-term dependency of portfolio assets and establish the dependency structure of the index returns of Shanghai and Shenzhen stock markets based on Copula function dependency model.Using the three-stage maximum likelihood The estimation method is used to estimate the parameters of the model, and the χ ~ 2 test statistic is used to test the model’s goodness-of-fit and model comparison.The results show that the model that considers the short-term dependency of individual assets in time is more suitable for describing the dependence of Shanghai and Shenzhen stock markets structure.