school

UM E-Theses Collection (澳門大學電子學位論文庫)

check Full Text
Title

An evaluation of credit risk scoring models in Macau banks

English Abstract

Credit Scoring models have been developed over several decades due to their increasingly important role in credit risk evaluation. The unpredictability and instability of the macroeconomics during Europe’s financial crisis has had a deep global impact on global economy, and the recovery is relatively slow. The financial institution’s sovereign rating of many developed countries has been lowered, and many international corporations have gone bankrupt. As such, it is the current trend for financial institutions to be concerned with how to control different kinds and levels of risk, including operation risk, market risk, credit risk, and so on. Risk control and management will be a crucial issue in the coming decades. This thesis focuses on how to predict credit risk and classification into different groups by Credit Scoring Models, which not only perform well on personal loans but also small and medium enterprise (SME) loans. In addition, the human adjustment method is still the mainstream in Macau, resulting in the credit risk evaluation depending on the credit analyst’s subjective perspective, experience and knowledge. It is believed that different standards exist between different banks and also within individual banks, resulting in differences at the operation and management levels and between credit analysts. How then do banks move forward instead of depending on human adjustment? Such a situation is similar to that of Egyptian banks. In order to ensure more efficient and effective credit risk evaluation methods of retail loans or SME loans, various researchers have applied credit scoring models to Egyptian banks’ credit risk evaluation. The choice of credit scoring models is because they are amongst the most widely used models in the world. Three traditional statistical techniques have been chosen and applied to this thesis, since their evaluated use in Egyptian banks is considered satisfactory. These methods are iii Discriminant analysis (DA), Logistic Regression (LR) and Probit analysis (PA). Through comparing the three models, the purpose of this thesis is to see whether these models are suitable for use in Macau’s banks and which model is most suitable for Macau’s current situation.

Issue date

2012.

Author

Leong, Pui Wa

Faculty

Faculty of Business Administration

Department

Department of Finance and Business Economics

Degree

M. Sc.

Subject

Banks and banking -- Macau

Credit scoring systems

Commercial loans

Risk management

Supervisor

Shu, Lian Jie

Files In This Item

TOC & Abstract

Full-text (Intranet only)

Location
1/F Zone C
Library URL
991001820609706306