【摘要】 本文以我国45家上市金融机构为样本，分别使用12种非对称和4种对称Copula模型拟合“系统—机构”二元相依结构，对比基于最优非对称和最优对称Copula-广义Co Va R的估计精度，并从宏观金融和微观机构层面分析影响金融机构系统性风险的重要因素。结果表明：时变非对称尾部相依是“系统—机构”相依关系的普遍特征；我国金融业各子部门的系统性风险排序基本为“银行>保险>证券>多元金融”；金融机构的个体风险与系统性风险的相关性较弱，下尾相依性是识别系统重要性金融机构的关键因素；在危机期间，金融机构的杠杆率对个体风险和系统性风险具有显著的正向影响；稳定和改善宏观金融环境是化解系统性风险的根本举措。更多还原

【Abstract】 This paper takes 45 Chinese listed financial institutions as a sample, using 12 asymmetric and 4 symmetric Copulas to fit the "system-institution" binary dependency structure, comparing the estimation accuracy of the generalized CoVaR based on optimal asymmetric and optimal symmetric Copulas, and analyzing the important factors that affect the systemic risk in the aspect of macro finance and micro institutions. The results show that time-varying asymmetric tail dependency is a common feature of the "system-institution" dependency relationship; The systemic importance ranking of various sub sectors of Chinese financial industry is basically "banking>insurance>broker>diversified finance"; The correlation between individual risk and systemic risk is weak, and lower tail dependency is a key factor in identifying systemically important financial institutions;During the crisis period, the leverage ratio of financial institutions has a significant positive impact on individual risk and systemic risk; Stabilizing and improving the macro financial environment is a fundamental measure to address systemic risk.更多还原