Application of Linear Logistic and Discriminant Analysis on Forecasting Creditworthiness of Individual Borrowers
View/ Open
Date
2013Author
Maranga, Bokea Samuel
Type
ThesisLanguage
enMetadata
Show full item recordAbstract
This research study summarizes the evaluation of credit risk using credit scoring method.
Credit scoring is a technique that helps banks decides whether to grant credit to applicants
who apply to them or not. The main objective of the research was to evaluate credit risk in
commercial banks using credit scoring models by ranking them based on their behavioural
financial and non- financial characteristics to honour their debt obligation in future. We
applied both logistic regression and discriminant analysis to identify predictors of default and
risk factors among cardholders followed for a period of eighteen months. A credit scoring
model was developed which can be used by commercial banks to determine the
creditworthiness of individual borrowers requesting for credit cards. The results showed out
that females constituted 64.3% of the population and they were the most disciplined. Type I
and type II errors had been calculated for all the credit scoring models used. The results
shows that the proposed model - Linear Logistic Model has more accuracy rate with less
misclassification cost errors as compared Discriminant Analysis. Also, several suggestions
for further research were presented.
Citation
Maranga Bokea Samuel (2013). Application of Linear Logistic and Discriminant Analysis on Forecasting Creditworthiness of Individual Borrowers. Masters of Science in Social StatisticsPublisher
University of Nairobi School of Mathematics