复杂金融网络中的风险传染与救助策略——基于中国金融无标度网络上的SIRS模型

发布时间：2018-06-07 12:19

本文选题：复杂网络 + 风险传染　；参考：《财贸经济》2017年04期

【摘要】：本文将传播动力模型SIRS引入到无标度的金融网络中,探讨了模型参数——感染率、治愈率、免疫失效率和网络紧密度对风险传染的影响。理论分析表明:具有无标度性的金融网络中,风险感染总是存在;风险传染会呈现"超调"现象,即感染比例会在短期内超越均衡值;危机中,增强金融机构的治愈能力比预防机构被感染和增强机构免疫能力的效果更好;减小机构之间的紧密性会降低危机的传染程度,但同时也延长了危机持续的时间。通过公开的中国大额支付系统数据,我们近似地构造了具有无标度特征的中国金融网络,并在此基础上进行了风险传染和救助的数值模拟。研究发现:采取救助措施时,多次适量救助是更优的策略,可以大幅降低危机峰值的"超调"现象;救助应该从危机加速度出现转折的时刻开始,从度大的机构逐渐向度小的机构过渡。这一救助思路对于特定网络结构下的系统风险控制具有一定的实践指导意义。
[Abstract]:In this paper, the transmission power model (SIRS) is introduced into the scale-free financial network, and the effects of the model parameters-infection rate, cure rate, immune failure rate and network tightness on risk transmission are discussed. Theoretical analysis shows that risk infection always exists in scale-free financial networks, and risk infection appears "overshoot" phenomenon, that is, the infection ratio will exceed the equilibrium value in a short period of time. Strengthening the healing capacity of financial institutions is better than preventing infection and enhancing institutional immunity; reducing the tightness between institutions will reduce the contagion of the crisis, but it will also prolong the duration of the crisis. Based on the open data of large payment system in China, we construct a scale-free Chinese financial network approximately, and then simulate the risk contagion and rescue. The study found that: when taking rescue measures, multiple times of appropriate rescue is a better strategy, can significantly reduce the crisis peak "overshoot" phenomenon; relief should begin at the moment of crisis acceleration turning point. The transition from a large mechanism to a small one. This idea has some practical significance for system risk control under specific network structure.
【作者单位】：中国社会科学院金融研究所;国家金融与发展实验室全球经济与金融研究中心;
【分类号】：F832

【相似文献】