金融机构信息网络与系统性风险——基于高频交易数据信息的分析  

Financial Institution Information Network and Systemic Risk——Analysis Based on High Frequency Trading Data Information

作　　者：宋晓彤 邢天才[2] 李孝溢 Song Xiaotong;Xing Tiancai;Li Xiaoyi

机构地区：[1]东北财经大学金融学院 [2]东北财经大学货币金融研究院 [3]中国人民银行大连市中心支行

出　　处：《投资研究》2022年第10期19-38,共20页Review of Investment Studies

基　　金：国家自然科学基金项目(71273042)。

摘　　要：金融市场是对信息高度依赖的市场,本文使用贝叶斯网络结构学习构建金融机构信息网络,考察信息网络特征对系统性风险的影响。研究发现:网络中心性与连通性越大导致系统性风险水平越高,银行与证券系统性风险对信息网络中心性变化更加敏感。2015年股市异常波动与2018年贸易摩擦期间,信息网络连通性与密度分别是影响两个时期金融机构系统性风险的主要网络特征。在平稳期与波动期,投资者情绪均为信息网络特征影响系统性风险的传导路径。The financial market is highly dependent on information.This paper uses Bayesian network structure to learn to construct the information network of financial institutions,and examines the influence of information network characteristics on systemic risk.The study found that the greater the network centrality and connectivity,the higher the level of systemic risk,and the systemic risk of banking and securities is more sensitive to changes in the centrality of the information network.During the abnormal volatility of the stock market in 2015 and the trade friction in 2018,the connectivity and density of information networks were the main network characteristics that affected the systemic risk of financial institutions in the two periods.In both the stable period and the fluctuating period,investor sentiment is the transmission path through which the characteristics of the information network affect the systemic risk.

关 键 词：信息网络 系统性风险 贝叶斯网络结构学习 投资者情绪 

分 类 号：F832[经济管理—金融学] TP311.13[自动化与计算机技术—计算机软件与理论]