基于DCC-GARCH模型的中国上市银行系统性风险研究

A Research into Systematic Risks of Listed Banks in China Based upon a DCC-GRACH Model

以我国15家上市银行为研究对象,就银行间的风险联动关系进行研究。研究主要包括:上市银行收益率两两间的时变相关系数测算、4家大型国有商业银行与11家股份制商业银行间的整体相关程度的测算,以及根据我国上市银行间的动态相关关系构建银行体系风险联动的预警指标。研究发现:我国上市银行间普遍存在显著的非对称的动态相关关系,4家大型国有银行间的平均动态相关系数比他们和其余11家银行间的相关系数高;我国的4家大型国有商业银行与11家股份制商业银行的整体相关程度也很高;15家上市银行两两的动态条件相关系数序列构建的银行体系系统性风险预警指标能够及时检测市场风险。

The article targets at 15 listed banks in China and analyses the risk correlation a- mong banks. The research on the interbank pertinence involves the following three parts： the measurement of time-varying correlation coefficients of yield rate of public banks; the measure- ment of the overall correlation between 4 large state-owned commercial banks and 11 joint-stock commercial banks; and the establishment of early-warning index of banking system risks correla- tion according to dynamic correlation among the listed banks in China. The study indicates that： a kind of unsymmetrical dynamic correlation prevails among the listed banks, in which the aver- age dynamic correlation coefficient of four large state-owned commercial banks is higher than that of eleven banks; the overall correlation between four large state-owned commercial banks and e- leven joint-stock commercial banks is also found high; the early-warning index of banking system risks correlation built on the basis of dynamic correlation index among fifteen listed banks can de- tect the market risks in time.