基于逻辑回归的个人信用评分卡模型研究
Research on Personal Credit Score Card Model Based on Logistic Regression
张俊丽 1郭双颜 2任翠萍 1马倩1
作者信息
- 1. 西安欧亚学院,陕西 西安 710065
- 2. 交通银行陕西省分行,陕西 西安 710004
- 折叠
摘要
构建有效的个人信用风险评价系统,用以应对潜在的个人信贷风险,这对金融行业和社会公众皆有重要的现实意义.文章首先对数据进行清洗、预处理,然后通过WOE编码分箱、IV值进行变量筛选,构建了逻辑回归模型并基于逻辑回归模型建立了个人信用评分卡模型,该模型可辅助决策者制定合理的授信政策、定价策略以及其他相关业务运营策略.
Abstract
Building an effective personal credit risk assessment system to address potential personal credit risks is of great practical significance for the financial industry and the general public.This paper first cleans and preprocesses the data,and then uses WOE coding and binning and IV values for variable screening.A logistic regression model is constructed,and a personal credit score card model is established based on the logistic regression model.This model can assist decision-makers in formulating reasonable credit policies,pricing strategies,and other related business operation strategies.
关键词
个人信用评估/评分卡/AUCKey words
personal credit evaluation/score card/AUC引用本文复制引用
基金项目
陕西省教育科学规划课题(十三五)(2020)(SGH20Y1480)
西安欧亚学院校级项目(2022)(2022GCPY01)
出版年
2024
国家科技期刊平台
NSTL
万方数据