论文部分内容阅读
引入SJC Copula模型来刻画中国房地产业与银行业在2000~2013年的的动态相依性,并进一步分析这种动态相依性的结构突变特征。研究结果表明:SJC Copula函数相比常见的其他Copula函数能更好地刻画房地产和银行业之间的动态相依结构;房地产业与银行业之间存在具有显著持续性的上(下)尾相依性,而且随着收益率的变化而呈现出非对称和非线性变化特征,主要表现为随着收益率的上升,条件上尾(下尾)相关系数将减小,但上尾相依性系数的下降幅度比下尾相依性系数要大,这说明房地产业和银行业之间的相依性的变化显著依赖于过去的收益的变化并且存在反向关系。房地产业与银行业之间的上(下)尾时变相关系数均具有“自相关、有偏、ARCH效应”的非正态分布特征,而且存在多个结构突变点;这些突变点发生日期附近往往伴绥着我国重大房地产金融调控政策的出台,具有明显的政策效应,但不具有持久性。
The SJC Copula model is introduced to describe the dynamic dependence of China’s real estate industry and banking industry from 2000 to 2013 and to further analyze the structural mutation characteristics of this dynamic dependence. The results show that the SJC Copula function can depict the dynamic dependency between real estate and banking better than other common Copula functions. There is a significant persistence of the upper (lower) tail dependency between real estate and banking , But also shows the characteristics of asymmetric and nonlinear changes with the change of return rate, which is mainly reflected by the decrease of the tail (lower tail) correlation coefficient as the rate of return increases, but the decreasing ratio of the upper tail dependence coefficient The lower end of the dependency coefficient is large, indicating that the dependence of the real estate industry and the banking industry depends significantly on changes in the past earnings and there is a reverse relationship. The upper and lower tail time correlation coefficients between real estate and banking all have non-normal distribution features of “autocorrelation, biased and ARCH effect ”, and there are multiple structural change points; these change points occur Near the date are often accompanied by the introduction of China’s major real estate finance regulatory policies, with obvious policy effects, but not durable.