信贷风险量化与信贷策略制定研究  

Credit Risk Quantification and Credit Strategy Formulation

作　　者：金昕怡 刁航 JIN Xin-yi;DIAO Hang(College of Mechanical and Electrical Engineering,Northeast Forestry University,Harbin 150040 China;College of Information and Computer Engineering,Northeast Forestry University,Harbin 150040 China)

机构地区：[1]东北林业大学机电工程学院,黑龙江哈尔滨150040 [2]东北林业大学信息与计算机工程学院,黑龙江哈尔滨150040

出　　处：《科技创新与生产力》2021年第7期61-64,共4页Sci-tech Innovation and Productivity

摘　　要：本文利用决策树模型对银行历史信贷数据进行学习,建立可以自动评估企业信誉等级的模型,结合发票的金额信息,建立CreditRisk+模型,通过计算各企业违约的概率,量化信贷风险的评价指标。结合新冠肺炎疫情严重影响经济与社会发展实际情况,采取博弈论的思想与基础模型对问题从社会、国家因素进行分析,最终得到适用于综合评价企业及未来信贷风险预测的模型,使银行的信贷策略更加贴近突发状况下的实际需要。This paper uses decision tree model to learn historical credit data of bank,establishes a model that can automatically evaluate enterprise credit rating.Combined with invoice amount information,it establishes a Creditrisk+model,quantifies credit risk evaluation indicators by calculating the probability of default by each enterprise.Combining the actual situation that COVID-19 seriously affects economic and social development,the game theory and basic models are adopted to analyze the problems from social and national factors,and finally a model suitable for comprehensive evaluation and credit risk prediction of enterprises is obtained,which makes the credit strategy of bank more closer to the actual needs in emergencies.

关 键 词：信贷风险预测 CREDITRISK+模型 决策树 博弈论 

分 类 号：F832.4[经济管理—金融学]