摘要
本文通过SSA-BP神经网络模型对系统性金融风险进行研究,并对中国的系统性金融风险进行评估和预警。第一,本文选取我国2008—2022年18个金融指标的月度数据构建了初始金融指标体系,在此基础上运用主成分分析和K-均值聚类将金融风险划分为四类。第二,基于SSA-BP神经网络模型建立我国金融风险预警模型,并通过2022年的数据对2023年的金融系统性风险状态进行仿真预测。结果显示,2023年的金融系统风险处于警戒状态或危险状态,值得重点关注。
In this paper,systematic financial risks are studied by SSA-BP neural network model,and systematic financial risks in China are assessed and warned.Firstly,the monthly data of 18 financial indicators in China from 2008 to 2022 are selected to construct the initial financial indicator system,based on which the financial risks are classified into four categories by using principal component analysis(PCA)and K-mean clustering.Secondly,the early warning model of China’s financial risks is established based on the SSA-BP neural network model,and the financial systemic risk status in 2023 is simulated and predicted by the data of 2022.The results show that the financial systemic risks in 2023 are in an alert or dangerous state,which is worth focusing on.
作者
陈庆婉
张品一
CHEN Qingwan;ZHANG Pinyi(Beijing Information Science&Technology University,Beijing 100192)
出处
《中国商论》
2023年第16期116-119,共4页
China Journal of Commerce
基金
“十四五”期间北京金融风险及防范对策研究(21JJB006)
北京市属高等学校优秀青年人才培育计划项目(BPHR202203240)。
关键词
SSA-BP神经网络模型
系统性金融风险预警
主成分分析
聚类分析
仿真预测
SSA-BP neural network model
early warning of systemic financial risks
principal component analysis(PCA)
cluster analysis
simulation prediction
