金融时间序列的短期相依性研究.pdf

2011年 03 月 M ar, 2011 ( 第25 卷第 03 期) East China Econom icM anagem ent (Vol. 25, No. 03 ) !DO I∀ 10 969/ jissn1007- 509720110 017 金融时间序列的短期相依性研究 易文德 (重庆文理学院 数学与统计学院, 重庆 402160) [] 金融资产相依结构的研究在金融风险分析中有着重要的意义金融资产的相依结构主要有两类: 一类 是单 个金融资产自身时间前后交易价格波动的相依关系, 称为短期相依关系, 另一类是金融资产间的价格波动 相依结构, 称为同期相依关系针对前 一种相依关系, 我们应用混合相依结构M - Copu la函数模型对上海综合 指数香港恒生指数和美国道琼斯指数三种金融时间序列前后一个交易日的价格波动相依关系进行了分析应 用两步骤法对模型的参数进行估计, 并对边缘分布和 M - Copula模型进行了拟合优度检验结果表明: 混合 M - Copu la模型能够捕捉金融资产时间序列的短期相依关系的变化规律 [] 短期相依; Copu la函数; 时间序列; 尾部相关 [] F8 0 [] A [] 1007 5097( 2011) 0 0071 05 Study on the TemporalDependence of Financial Tim e Series Y IW ende (S chool of M athem atics S tatist ics, Chong q ing Un iv ersity of A rts and Sc iences, Chong qing 402160, Ch ina ) Abstract: It is greatly in teresting to investigate the dependence structure of financial assets in financial risk analysis. There are two types of dependence structures of financial assets: one is the dependence relationship of individual financial asset itself in different tmi e which is called tem poral dependence relationship and the other is the dependence structure between different financial assetsw hich is defined as contem poraneous dependence. In th is paper, w e focus on the form er and propose aM - Copulam odel to investigate the tem poral depend ence for th ree stock m arkets: the ShanghaiCom posite Index ( SH), theH ang Seng Index ( HK) and Dow - Jones Index ( DJ). The wt o- stage maxmi um likelihood estmi ation is em ployed to estmi ate the param eters of m odel and the goodness of fit of margins and M - Co