Data Driven Optimal Combination Evaluation Model of Enterprise Credit Risk

doi:10.3778/j.issn.1002-8331.2202-0261

Abstract

Abstract: Various types of information provided by corporate financial data can effectively explain the credit level of the company. However, too many indicators often have multicollinearity problems, resulting in the model overfitting and reducing the evaluation accuracy. In order to reduce the indicator data, the correlation coefficient between corporate default and financial indicators is calculated to eliminate indicators with weak correlation. The Lasso regression method is used to reduce the index data with high correlation. Then, three classification models of Logistic regression model, Bayesian model and support vector machine are used to classify and evaluate the credit risk of enterprises. Considering the different classification accuracy of different methods for different enterprises, in order to comprehensively utilize the advantages of each method, an optimal combination evaluation model of enterprise credit risk based on integer programming is constructed. Finally, the simulation analysis is carried out on the data of 300 companies listed on the Growth Enterprise Market. In order to verify the validity of the model, this paper randomly selects 3 groups of samples from the 300 companies（270 companies are training samples and 30 companies are test samples）. The data of one year before the special treatment（ST） is performed are tested, and the experimental results show that the combined model has higher stability and classification accuracy.

Key words: Lasso regression, optimal combination evaluation, Logistic regression, Bayesian classification, support vector machine（SVM）

摘要： 企业财务数据所提供的各类信息可以有效解释企业的信用水平，然而过多指标往往存在多重共线性问题，造成模型的过拟合，反而降低评价精度。为了约简指标数据，计算企业违约情况与财务指标的相关系数，剔除掉相关性弱的指标。采用Lasso回归方法对相关性高的指标数据进行约简，进而利用Logistic回归模型、贝叶斯模型和支持向量机三种分类模型对企业的信用风险进行分类评价。考虑到不同方法对不同企业的分类精度不同，为了综合利用各方法的优势，构建基于整数规划的企业信用风险最优组合评价模型。对300家创业板上市企业数据进行仿真分析，为了验证模型的有效性，在300家公司中（其中270家为训练样本，30家为测试样本）随机选取3组样本，使用ST公司被执行特别处理（special treatment，ST）前一年的数据进行测试，实验结果表明组合模型具有更高的稳定性和分类精度。

关键词: Lasso回归, 最优组合评价, Logistic回归, 贝叶斯分类, 支持向量机（SVM）

LUO Min, ZHOU Ligang, LIU Xinyue, ZHU Jiaming, CHEN Huayou. Data Driven Optimal Combination Evaluation Model of Enterprise Credit Risk[J]. Computer Engineering and Applications, 2023, 59(10): 306-313.

罗敏, 周礼刚, 刘欣悦, 朱家明, 陈华友. 数据驱动的企业信用风险最优组合评价模型[J]. 计算机工程与应用, 2023, 59(10): 306-313.

References

[1] BEAVER W.Financial ratios as predictors of failures，empirical research in accounting：selected studies[J].Journal of Accounting Research，1966（4）：71-111．
[2] ALTMAN E I.Financial ratios，discriminant analysis and the prediction of corporate bankruptcy[J].Journal of Finance，1968，23（4）：589-609.
[3] MASMOUDI K，ABID L，MASMOUDI A.Credit risk modeling using Bayesian network with a latent variable[J].Expert Systems with Applications，2019，127（8）：157-166.
[4] 陈诗一.德国公司违约概率预测及其对我国信用风险管理的启示[J].金融研究，2008（8）：53-71.
CHEN S Y.Prediction of default probability of German companies and its enlightenment to China’s credit risk management[J].Journal of Financial Research，2008（8）：53-71.
[5] 邬建平.基于粗糙集和支持向量回归的电子商务信用风险评估[J].统计与决策，2019，35（23）：53-56.
WU J P.E-commerce credit risk assessment based on rough set and support vector regression[J].Statistics & Decision，2019，35（23）：53-56.
[6] 梁龙跃，刘波.基于文本挖掘的上市公司财务风险预警研究[J].计算机工程与应用，2022，58（4）：255-266.
LIANG L Y，LIU B.Research on financial risk early warning of listed companies based on text mining[J].Computer Engineering and Applications，2022，58（4）：255-266.
[7] SHEN F，ZHAO X，KOU G，et al.A new deep learning ensemble credit risk evaluation model with an improved synthetic minority oversampling technique[J].Applied Soft Computing，2021，98（1）：106852.
[8] WANG F，DING L，YU H，et al.Big data analytics on enterprise credit risk evaluation of e-business platform[J].Information Systems and e-Business Management，2020，18（3）：311-350.
[9] ELENA D，SULLIVAN H，CHRISTOPHE H，et al.Machine learning for credit scoring：improving logistic regression with non-linear decision-tree effects[J].European Journal of Operational Research，2022，297（3）：1178-1192.
[10] 彭伟.基于随机森林-Logistic回归组合模型的小额贷款用户逾期预测[D].重庆：重庆大学，2019.
PENG W.Overdue prediction of microfinance users based on random forest logistic regression combination model[D].Chongqing：Chongqing University，2019.
[11] 陈云，石松，潘彦，等.基于SVM混合集成的信用风险评估模型[J].计算机工程与应用，2016，52（4）：115-120.
CHEN Y，SHI S，PAN Y，et al.Hybrid ensemble approach for credit risk assessment based on SVM[J].Computer Engineering and Applications，2016，52（4）：115-120.
[12] DAHOOIE J H，HAJIAGHA S，FARAZMEHR S，et al.A novel dynamic credit risk evaluation method using data envelopment analysis with common weights and combination of multi-attribute decision-making methods[J].Computers & Operations Research，2021，129（1）：105223.
[13] SHAFIEE S，LIED L M，BURUD I，et al.Sequential forward selection and support vector regression in comparison to LASSO regression for spring wheat yield prediction based on UAV imagery[J].Computers and Electronics in Agriculture，2021，183（4）：106036.
[14] ARORA N，KAUR P D.A Bolasso based consistent feature selection enabled random forest classification algorithm：an application to credit risk assessment[J].Applied Soft Computing，2020，86：105936.
[15] MCCALLUM A，NIGAM K.A comparison of event models for naive Bayes text classification[C]//AAAI-98 Workshop on Learning for Text Categorization，1998：41-48.
[16] VAPNIK V.SVM method of estimating density，conditional probability，and conditional density[C]//2000 IEEE International Symposium on Circuits and Systems，Geneva，May 28-30，2000：749-752.
[17] 肖斌卿，柏巍，姚瑶，等.基于LS-SVM的小微企业信用评估研究[J].审计与经济研究，2016，31（6）：102-111.
XIAO B Q，BO W，YAO Y，et al.Small and micro businesses credit evaluation research based on LS-SVM[J].Audit & Economy Research，2016，31（6）：102-111.
[18] 查璐瑶.创业板上市公司财务风险评价研究[D].合肥：合肥工业大学，2017.
ZHA L Y.Research on financial risk evaluation of company listed on GEM[D].Hefei：Hefei University of Technology，2017.