The profitability of online loans: A competing risks analysis on default and prepayment

Zhiyong Li, Aimin Li, Anthony Bellotti, Xiao Yao

Research output: Journal PublicationArticlepeer-review

Abstract

Traditional credit scoring models help lenders to make informed decisions in identifying those borrowers most likely to default. We analyse over one million online loans and find that the rates for both default and prepayment are relatively high compared to traditional bank loans. A preliminary nonparametric life-table estimate shows that loans with different terms exhibit varying patterns of hazards. We use a proportional hazard model with competing risks to predict the time to default and prepayment, and parameterise those covariates affecting the time to both events. Two dimensions of predictive performance, the discriminant power and the probability calibration, are then examined. To further support the primacy of profit-driven decisions, we propose a framework based on competing risks survival analysis to estimate the profitability of loans and the return of loan portfolios. Profitability forecasts incorporating both the time to default and prepayment are compared to the profitability of real assets, and finally a penalty is suggested to compensate for those losses incurred by prepayment.

Original languageEnglish
JournalEuropean Journal of Operational Research
DOIs
Publication statusAccepted/In press - 2022

Keywords

  • Competing risks
  • Credit scoring
  • OR in banking
  • Profitability
  • Survival analysis

ASJC Scopus subject areas

  • Computer Science (all)
  • Modelling and Simulation
  • Management Science and Operations Research
  • Information Systems and Management
  • Industrial and Manufacturing Engineering

Fingerprint

Dive into the research topics of 'The profitability of online loans: A competing risks analysis on default and prepayment'. Together they form a unique fingerprint.

Cite this