Credit risk modelling in peer-to-peer lending: a comparative analysis of neural networks and XQboost
Date
2021
Authors
Wachira, Njomo
Journal Title
Journal ISSN
Volume Title
Publisher
Strathmore University
Abstract
Consumer credit risk modelling involves coming up with the probability that a borrower default thus classifying borrowers as either defaulters or non-defaulters. This is important for lending firms because they are then able to lend out cash to borrowers who are most likely to repay on time. Tl1is protects their profits. This study aims to use machine learning techniques I in consumer credit risk modelling in a bid to find out which technique is more effective. In addition, the study aims at investigating the effect of new credit customers on default expenence.
Description
Submitted in partial fulfilment of the requirements for the Degree of Bachelor of Business Science in Actuarial Science at Strathmore University