Credit score in default prediction for p2p lending

Sim, Hui Xian (2024) Credit score in default prediction for p2p lending. Final Year Project, UTAR.

Preview

PDF Download (733Kb) | Preview

Abstract

This research aims to investigate the factors and criteria influencing default in Peer to Peer (P2P) lending, with a focus on providing valuable insights for the future of Fintech and contributing to industry growth and sustainability. The study examines the intention to adopt P2P lending and its implications on financial decision-making. Utilizing quantitative methods, the analysis incorporates variables including loan amount, interest rate, total open-to-buy on revolving bank cards, bank card utilization rate, number of open revolving accounts, debt to income ratio (DTI), and revolving utilization rate. Data from a Kaggle dataset for the year 2018 comprising 445 samples with charge-offs, late payments of 16-30 days, and late payments of 31-120 days is analyzed. Results indicate a highly positive relationship with revolving utilization rate and negative relationships with the number of open revolving accounts and total open-to-buy on revolving bank cards. The implications suggest enhancing credit monitoring within credit assessment processes and implementing alternative data for more accurate evaluations. By accessing FICO scores to assess creditworthiness based on consumer payment behaviour is recommended to improve loan approval processes

Actions (login required)