基于零膨胀分位数两部模型的银行贷款违约预测研究

doi:10.16381/j.cnki.issn1003-207x.2020.0441

Abstract

Abstract: Loan is not only a main means of solving the shortage of finances, but also an important business of financial institutions. Loan default forecasting is an essential content of bank risk management. To measure the loan credit risk of lenders, the number of days overdue is an informative variable commonly used. It shows whether the lender defaulted or not, but also the extent of default. However, this variable usually has an obvious zero-inflated characteristic, that is, there exists a quite large proportion of zero observations. Those zeros usually bring challenges to traditional credit default forecasting models.

Key words: bank loan; two-part model; penalized variable selection; quantile regression

CLC Number:

F830.589

WANG Xiao-yan, YUAN Teng, DUAN Xiang-bin. Loan Default Forecasting Based on Zero-inflated Quantile Two-part Model[J]. Chinese Journal of Management Science, 2022, 30(10): 1-13.

References

［1］ David H S, Jon S H. Predicting national direct student loan defaults: The role of personality data［J］. Research in Higher Education, 1979, 10(3): 195-205.
［2］ Foster C C, Order R V. An option based model of mortgage default［J］. Housing Finance Review, 1984, 3(4): 351-359.
［3］ Ma Ruowei, Yang Deyong. Building default predicting model on firm's short-term loan data with artificial neural network - considering qualitative indexes and misclassification costs［C］//International Conference on E-business and E-government, Guangzhou, China, May 7-9, 2010.
［4］杨保安, 季海. 基于人工神经网络的商业银行贷款风险预警研究［J］. 系统工程理论与实践, 2001, 21(5): 70-74.Yang Baoan, Ji Hai. A study of commercial bank loans risk early warning based on BP neural network［J］. Systems Engineering-Theory and Practice, 2001, 21(5): 70-74.
［5］方匡南, 吴见彬, 朱建平, 等. 信贷信息不对称下的信用卡信用风险研究［J］. 经济研究, 2010, 45(S1): 97-107.Fang Kuangnan, Wu Jianbin, Zhu Jianping, etal. Forecasting of credit card credit risk under asymmetric information based on nonparametric random forests［J］. Economic Research Journal,2010,45(S1):97-107.
［6］ Danenas P, Garsva G. Selection of support vector machines based classifiers for credit risk domain［J］. Expert Systems with Applications, 2015, 42(6): 3194-3204.
［7］ Lobna A, Afif M, Sonia Z. The consumer loan’s payment default predictive model: An application of the logistic regression and the discriminant analysis in a Tunisian commercial bank［J］. Journal of the Knowledge Economy, 2018, 9(3): 948-962.
［8］胡毅, 王珏, 杨晓光. 基于面板Logit模型的银行客户贷款违约风险预警研究［J］. 系统工程理论与实践, 2015, 35(7): 1752-1759.HuYi, Wang Jue, Yang Xiaoguang.A study on early-warning of bank customer loan defaults based on panel Logit model［J］. Systems Engineering-Theory and Practice, 2015, 35(7): 1752-1759.
［9］舒扬, 杨秋怡. 基于大样本数据模型的汽车贷款违约预测研究［J］. 管理评论, 2017, 29(9): 59-71.Shu Yang,Yang Qiuyi.Research on auto loan default prediction based on large sample data model［J］.Business Review, 2017, 29(9): 59-71.
［10］石庆焱. 一个基于神经网络—Logistic回归的混合两阶段个人信用评分模型研究［J］. 统计研究, 2005, 22(5): 45-49.Shi Qingyan.A mixed two-stage personal credit scoring model based on neural network logistic regression［J］.Statistical Research, 2005, 22(5): 45-49.
［11］张奇, 胡蓝艺, 王珏. 基于Logit与SVM的银行业信用风险预警模型研究［J］. 系统工程理论与实践, 2015, 35(7): 1784-1790.Zhang Qi, Hu Lanyi, Wang Jue.Study on credit risk early warning based on Logit and SVM［J］.Systems Engineering-Theory and Practice, 2015, 35(7): 1784-1790.
［12］陈林, 谢彦妩, 李平, 等.借款陈述文字中的违约信号—基于 P2P 网络借贷的实证研究［J］. 中国管理科学, 2019, 27(4): 37-47.Chen Lin, Xie Yanwu, Li Ping, et al. The signal of default risk from the description-text based on the empirical research of P2P lending［J］.Chinese Journal of Management Science, 2019, 27(4): 37-47.
［13］王小燕, 张中艳, 马双鸽. 基于文本先验信息的贷款信用风险评估模型［J］.中国管理科学, 2021, 29(5): 34-44.Wang Xiaoyan, Zhang Zhongyan, Ma Shuangge. A loan credit risk model incorporating text prior information［J］. Chinese Journal of Management Science, 2021, 29(5): 34-44.
［14］王呈斌, 方匡南, 郑陈璐. 基于随机生存森林的房屋贷款逾期研究［J］. 上海金融, 2020(2): 59-63.Wang Chengbin, Fang Kuangnan, Zheng Chenlu. Research on overdue housing loan based on random survival forest［J］. Shanghai Finance, 2020(2): 59-63.
［15］程秀媚. 区域宏观经济因素对P2P借款人逾期时间的影响［D］. 北京: 对外经济贸易大学, 2015.Cheng Xiumei. Influence of regional macroeconomic factors on overdue time of repayment from P2P borrowers［D］. Beijing: University of International Business and Economics, 2015.
［16］ Lambert D. Zero-inflated Poisson regression, with an application to defects in manufacturing［J］. Technometrics, 1992, 34: 1-14.
［17］ Diaz J, Joseph M B. Predicting property damage from tornadoes with zero-inflated neural networks［J］. Weather and Climate Extremes, 2019, 25: 1-16.
［18］姜炜. 零膨胀两部模型在贷款违约风险因素分析中的应用［D］. 长沙: 湖南大学, 2018.Jiang Wei.The application of zero-inflated two-part model in the analysis of the risk factors of loan default［D］. Changsha:Hunan University, 2018.
［19］何晓群, 夏利宇, 姜天英. 处理不平衡征信数据的零膨胀信用评级模型［J］.数理统计与管理, 2019, 38(5): 812-822.He Xiaoqun, Xia Liyu, Jiang Tianying.Zero-inflated credit score model based on imbalance dataset［J］.Journal of Applied Statistics and Management, 2019, 38(5): 812-822.
［20］孟生旺, 杨亮. 随机效应零膨胀索赔次数回归模型［J］. 统计研究, 2015(11): 97-102.Meng Shengwang, Yang Liang.Random-effect zero-inflated claims frequency regression models［J］.Statistical Research, 2015(11): 97-102.
［21］徐昕, 袁卫, 孟生旺. 零膨胀负二项回归模型的推广与费率厘定［J］. 系统工程理论与实践, 2012, 32(1): 127-133.Xu Xin, Yuan Wei, Meng Shengwang. Generalization of zero-inflated negative binomial regression model and rate making［J］. Systems Engineering-Theory and Practice, 2012, 32(1): 127-133.
［22］ Fang Kuangnan, Wang Xiaoyan, Shia Benchang, et al. Identification of proportionality structure with two-part models using penalization［J］. Computational Statistics and Data Analysis, 2016, 99: 12-24.
［23］夏丽丽, 田茂再. 零膨胀泊松回归模型的非参数统计分析及其应用［J］. 数理统计与管理, 2019,38(2): 235-246.Xia Lili, Tian Maozai.Nonparametric statistical analysis of zero-and-one inflated Poisson regression models and its application［J］.Journal of Applied Statistics and Management, 2019,38(2): 235-246.
［24］胡亚南, 田茂再. 零膨胀计数数据的联合建模及变量选择［J］. 统计研究, 2019, 36(1): 104-114.Hu Yanan, Tian Maozai.Joint modeling and variable selection from zero-inflated count data［J］. Statistical Research, 2019, 36(1): 104-114.
［25］夏丽丽, 田茂再. 一类带约束的零膨胀广义可加模型的惩罚似然估计［J］. 统计与决策, 2020, 36(2): 16-20.Xia Lili, Tian Maozai.Penalty likelihood estimation for a class of constrained zero-inflated generalized additive model［J］.Statistics and Decision, 2020, 36(2): 16-20.
［26］ Tibshirani R. Regression shrinkage and selection via the Lasso［J］. Journal of the Royal Statistical Society. Series B (Methodological), 1996, 58: 267-288.
［27］王小燕, 谢邦昌, 马双鸽, 等. 高维数据下群组变量选择的惩罚方法综述［J］. 数理统计与管理, 2015, 34(6): 978-988.Wang Xiaoyan, Xie Benchang, Ma Shuangge, et al.A review of penalized group variable selection methods in high dimensional data［J］.Journal of Applied Statistics and Management, 2015, 34(6): 978-988.
［28］晏翔. 商业银行个人贷款风险因素实证分析［J］.农村金融研究, 2018(5): 47-49.Yan Xiang.Risk factors of commercial bank retail loans［J］. Rural Finance Research, 2018(5): 47-49.
［29］程萍. 影响个人贷款违约风险因素的实证研究-以大众汽车金融公司为例［D］. 北京: 对外经济贸易大学, 2018.Cheng Ping. The empirical study of the default risk on individual loans of Volkswagen financial services company［D］. Beijing: University of International Business and Economics, 2018.

[1]	CHEN Lin, XIE Yan-wu, LI Ping, LI Qiang. The Signal of Default Risk from the Description-text Based on the Empirical Research of P2P Lending [J]. Chinese Journal of Management Science, 2019, 27(4): 37-47.
[2]	SHI Bao-feng, WANG Jing, CHI Guo-tai. The Inclusive Finance, Bank Loans and Financing of Small Private Business Microfinance Loan-Based on Matching Credit Risk and Credit Rating [J]. Chinese Journal of Management Science, 2017, 25(9): 28-36.

Loan Default Forecasting Based on Zero-inflated Quantile Two-part Model

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 2

Metrics

Comments

Recommended 10