The Role of Supply Chain Partner Information in Credit Risk Prediction of Supply Chain Finance——A Comparative Analysis Based on Multiple Machine Learning Models
Supplychainfinance(SCF)servesasaneffectivemeansforsmallandmedium-sized enterprises(SMEs)to alleviate financing difficulties.To address challenges in predicting credit risks in SCF,such as information asymmetry and sample selection bias,this study incorporates supply chain partner information into the risk indicator system.Utilizing disclosed data from A-share listed companies between 2010 and 2021,a comparative analysis is conductedby by using four machine learning models:random forest,XGBoost,logistic regression,and MLP.The results demonstrate that the inclusion of partner information enhances the accuracy and stability of credit risk prediction in SCF.Finally,through Lime interpretability analysis,six indicators,including partner's quick ratio,sales profit margin,and asset-liability ratio,are identified as the primary influencing factors for credit risk prediction in supply chain partner information.
NETL
NSTL
万方数据
