GBDT-SVM Credit Risk Assessment Model and Empirical Analysis of Peer-to-Peer Borrowers under Consideration of Audit Information

Zhou Li

doi:10.4236/ojbm.2018.62026

Abstract

With the rapid development of P2P (peer-to-peer) online lending industry, how to effectively evaluate the borrowers’ credit risk in the platform has drawn more and more attention. In this paper, we propose a borrower credit risk assessment index system that includes basic information, work information, credit information, asset information, loan information and audit certification information, and come up with a credit risk assessment model that combines Gradient Boosting Decision Trees (GBDT) and support vector machine (SVM). Then, we select the data of P2P lending platform to carry out the empirical analysis of the credit risk assessment, and compare with the common four kinds of single prediction models such as logic regression (LR), artificial neural network (ANN), SVM and clustering algorithm. The results show that the increase of audit certification information helps to improve the forecasting effect of the model, and the credit risk assessment model of P2P lending platform based on GBDT and SVM has higher prediction accuracy and stability.

Highlights

As a new financing model, Peer-to-Peer Lending rapidly expanded and aroused widespread concern in the social community
The results show that the increase of audit certification information helps to improve the forecasting effect of the model, and the credit risk assessment model of P2P lending platform based on Gradient Boosting Decision Trees (GBDT) and Support Vector Machine (SVM) has higher prediction accuracy and stability
The comparison shows that the comprehensive prediction results of logic regression (LR), artificial neural network (ANN) and GBDT-SVM models are relatively good, while the prediction results of clustering and SVM are poor

Summary

Introduction

As a new financing model, Peer-to-Peer Lending rapidly expanded and aroused widespread concern in the social community. According to the “P2P Lending Industry Report of the Whole Country in First Half Year 2016” published by Diyiwangdai platform, there were 2077 platforms that stopped operating, cash withdrawal difficulties, runways and other issues, with an increase of 559 over 2015. This will undoubtedly make investors and institutions face huge risks and will gradually reduce people’s trust in P2P lending, which will hinder the development of P2P lending industry. It can be concluded from the above facts, that to find out how to accurately evaluate the credit risk of P2P borrowers is important

Objectives

Results

Conclusion