Modelling Default Risk of Borrowers: Evidence from Online Peer to Peer Lending Platforms in Australia

The purpose of this research is to evaluate several popular machine learning algorithms for credit scoring for peer to peer lending. The dataset to fit the models is extracted from the official site of Lending Club. Several models have been implemented, including single classifiers (logistic regression, decision tree, multilayer perceptron), homogeneous ensembles (XGBoost, GBM, Random Forest) and heterogeneous ensemble classifiers like Stacked Ensembles. Results show that ensemble classifiers outperform single ones with Stacked Ensemble and XGBoost being the leaders.

Download Full-text

Evaluating borrower’s default risk in peer-to-peer lending: evidence from a lending platform in China

Applied Economics ◽

10.1080/00036846.2016.1262526 ◽

2016 ◽

Vol 49 (35) ◽

pp. 3538-3545 ◽

Cited By ~ 43

Author(s):

Xuchen Lin ◽

Xiaolong Li ◽

Zhong Zheng

Keyword(s):

Default Risk ◽

Peer To Peer ◽

Peer Lending

Download Full-text

Gender effect on the default risk in peer-to-peer lending markets: The case of the largest Chinese platform

Risk Governance and Control Financial Markets & Institutions ◽

10.22495/rgcv9i3p1 ◽

2019 ◽

Vol 9 (3) ◽

pp. 8-22

Author(s):

Lin Lingnan

Keyword(s):

Credit Risk ◽

Gender Discrimination ◽

Default Risk ◽

Probit Model ◽

Peer To Peer ◽

Gender Effect ◽

Individual Characteristics ◽

Model Estimate ◽

P2p Lending ◽

Peer Lending

Research of gender effect on funding success in peer-to-peer lending markets demonstrates that gender discrimination is a platform-specific phenomenon rather than a common feature. Can we get a similar conclusion about the relationship between gender and credit risk? How do gender differences affect default risk? We try to answer this question using the data of the largest peer-to-peer lending platform RenRenDai spanning from March 2016 to September 2016. In order to avoid the endogeneity problem, this paper first uses the instrumental variable method to conduct a baseline Probit model estimate connecting gender difference to the default rate with several borrowers’ individual characteristics under control. Then the original Probit model and a propensity score matching method aiming to eliminate the effects of divergent observable characteristics are applied to test the robustness of the outcome. Both the baseline estimation and the robustness test show that there is no significant gender effect on the probability of default, ceteris paribus. Therefore, borrowers’ gender is not a good screening device for the P2P lending platform to control the credit risk; other factors should be taken into account to reduce the non-performing loan rate. However, since this paper only investigates the situation of RenRenDai and the data we use is limited, we should be very careful to generalize our findings to other P2P lending platforms. More research on different P2P lending platforms in different regulatory regimes is in necessity

Download Full-text