基于合成数据的系统性风险传染研究

发布时间：2019-03-02 11:53

【摘要】：本文基于合成数据的方法和复杂网络理论,对银行间系统性风险的传染问题进行了研究。首先在只能获取支付系统和银行年度报表数据的约束下,采用合成数据的方法,估算出我国支付系统中银行间的交易数据,从而得到交易矩阵。以此合成数据为基础,进一步借助复杂网络理论,利用计算机仿真技术对系统性风险传染问题进行了研究。研究结果表明,国有商业银行抵御风险能力最强,城商行和农商行这类小型银行对国有商业银行有着很强的依赖性。最后形成了若干政策建议。以上成果有可能为金融风险传染理论的发展和宏观审慎管理制度的完善等起到一定的推动作用。
[Abstract]:Based on the method of synthetic data and the theory of complex network, this paper studies the contagion of systemic risk between banks. First of all, under the constraint of obtaining payment system and bank annual report data, the transaction data between banks in China's payment system is estimated by using synthetic data method, and then the transaction matrix is obtained. Based on the synthetic data and with the help of the complex network theory, the systemic risk contagion problem is studied by computer simulation technology. The results show that state-owned commercial banks have the strongest ability to resist risks, and small banks such as city commercial banks and agricultural commercial banks have strong dependence on state-owned commercial banks. Finally, a number of policy recommendations have been formed. These achievements may play an important role in the development of financial risk contagion theory and the improvement of macro-prudential management system.
【作者单位】：西南财经大学金融工程与金融智能四川省重点实验室;西南财经大学经济信息学院;
【基金】：国家社会科学基金青年项目(13CJY121) 中央高校基本科研业务费博士研究生课题项目(JBK1307151)的资助
【分类号】：F832.33

【参考文献】