为捕捉金融资产收益率尖峰厚尾和非对称特征,本文首次在广义双曲线分布族下对我国银行系统性风险进行动态化测量,并对条件风险价值方法在我国银行系统性风险测量的适用性进行检验。通过返回检验表明,t分布下的动态系统性风险测度未能达到准确标准,双曲线分布下的条件风险价值相对其他分布能更准确的刻画我国银行系统性风险;通过对各银行△CoVaR排序发现,国有银行系统性风险贡献高于股份制商业银行,但各银行的系统重要性排名会随着时间不断发生改变。
In order to better models the characteristics of skewness and fatness of financial asset returns, this paper first extends the CoVaR method with using the generalized hyperbolic distribution to measure the dynamic systemic risk of bank system. We also construct the back-test framework for the CoVaR method. The empirical results shows that the CoVaR based on t distribution fail to achieve the accurate standard and the estimated effect of hyperbolic distribution is better than the other distributions. We also find that the contribution to the systemic risk of the nationalized banks were higher than the stock- holding banks, while the monitoring of the systemic importance should be dynamic since it changed with time.