An unsupervised credit scoring system for unsecured loan lending
Date
2014
Authors
Kinuthia, Steven
Journal Title
Journal ISSN
Volume Title
Publisher
Strathmore University
Abstract
Due to growmg competition, over-indebtedness, and economic crises,
banks have to pursue their social and financial objectives in an increasingly
constrained environment. The key determinant of Survival, competition and
profitability is the quality of bank loans they give to their customer.
Consequently, developing powerful risk management tools becomes more than
ever crucial to survive. Credit scoring has been discussed in this study as one of
the most important kits to classify loan applicants as "good" or "bad"
The study also discusses how a credit scoring system can be developed to remove
redundancies and biases of credit officers in the credit scoring process, mitigate
the risk of lender to borrower and reduce the turnaround time of loan
application. It discusses the best techniques and key determinants in developing
such a systems and factors that should be considered to guarantee the accuracy
of the system.
A mixed method research design was used to investigate and review the
various credit evaluation techniques and to also analyze various credit scoring
techniques that may be used to develop the unsupervised credit scoring system.
Discriminant analysis was the preferred technique in development of the credit
scoring system due to its ease of use and implementation. Literature reviewed
also discusses it as one of the most widely used technique.
The research findings showed that even though credit scoring tools were
in place, the reliance on them was very low. Age, marital status, education and
history of the applicant have been discussed as some of the most important
factors in developing a credit scoring system.
Description
A Thesis submitted in partial fulfillment of the requirements for the award of the
Degree of Masters of Science in Information Technology
Keywords
Credit scoring system, Banking, Unsecured loan, ICT, Kenya