An Empirical Study on Prediction of the Default Risk on P2P Lending Platform

2nd International Symposium on Application of Materials Science and Energy Materials

材料科学;能源学

Qian, Meng^1 ; Hu, Fangqin^1

School of Computer and Imformation, Anqing Normal University, Anqing, China^1

关键词: Corresponding measures;    Empirical studies;    Feed-Forward;    Least squares support vector machines;    Loan default;    Logistic Regression modeling;    Optimization algorithms;    Transaction data;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/490/6/062048/pdf
DOI  :  10.1088/1757-899X/490/6/062048

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

P2P lending platform contains many risks of loan default. The top priority of P2P lending platform is to predict the risk of default accurately and to take corresponding measures. The paper selects public lending loan transaction data of users from Lending Club to carry on the empirical study on the risks of loan default by respectively using logistic regression model, the bat optimization algorithm of feedforward(BAPA) neural network and least squares support vector machine(LSSVM) to analyze the experimental data, and then to evaluate the applicability of three methods in predicting the risk of loan default on the P2P network platform. The experimental results show that the least squares support vector machine has a better prediction effect.