Abstract
Under Basel Capital Accord requirement, banks have come to the firm conclusion that they need to start quantifying credit risk. There are two main types of current credit risk assessment model, which are judgmental and quantitative. Judgmental type method is based on analyses relying on quantitative techniques and subjective elements. This method suffers a few disadvantages. Firstly, it is expensive in terms of resources and costs. Secondly, the transformation of tacit knowledge about the borrowers' characteristics to explicit organizational knowledge is difficult under such a process (Chakrabati, Baijayanta and Ravi Varadachari ). The quantitative method depends on statistical analysis and consumer historical data to arrive at the conclusion. The current quantitative model which actively being used by most banks for credit risk estimation is credit scoring model. The main objective of this study is to analyze the performance of credit scoring model to predict the credit loss based on Citibank's consumer data using the statistical method which is linear regression model. This performance accuracy was determined by representing the analysis of experiment's result. In addition to that, this study identifies the issues of weaknesses of current credit risk model in estimating the credit loss and issues of data limitation in implementing credit card's risk model.
Metadata
Item Type: | Thesis (Masters) |
---|---|
Creators: | Creators Email / ID Num. Zakaria, Norzaila Era UNSPECIFIED |
Contributors: | Contribution Name Email / ID Num. Thesis advisor Nordin, Ariza UNSPECIFIED |
Divisions: | Universiti Teknologi MARA, Shah Alam > Faculty of Computer and Mathematical Sciences |
Programme: | Master of Science |
Keywords: | Credit. Card, Bank |
Date: | 2006 |
URI: | https://ir.uitm.edu.my/id/eprint/64028 |
Download
64028.PDF
Download (13kB)