| Summary: | 碩士 === 國立高雄應用科技大學 === 金融資訊研究所 === 98 === Mortgage loan is a quite important credit business in bank industry. There are too many competitors in domestic financial industry, so it gradually makes the profit margins of mortgage loan narrow and the quality of credit low. When the rate of overdue loan is skyrocketing, it will endanger the operation of banks. Therefore, bank’s profit rely more on should control the risk management. If they are able to sort out the high-risk cases through using Logistic Regression and to prevent better than cure, they will not only get more profits but also avoid causing the social and economic problems.
This study collects mortgage loan cases in a branch of a southern Taiwan bank during 2004 to 2008 as samples. According to the number and scale of the mortgage loans, this bank within the top ranks can be the typical research samples. This research selected 341 cases in total; 268 cases of samples are regular ones that repay on time and 73 cases are overdue cases. The study investigates and analyzes the level of influences about the overdue mortgage loans with nine variables which include marital status, educational level, occupation, credit line, loan with grace period, credit loan, ratio of income and expenses, ratio of loan approval, and house age.
Based on the results of this study, the researcher concludes that marital status, educational level, loan with grace period, credit loan, ratio of income and expenses, and house age are the six significant variables which influence consumer loan.
The results indicate that the prediction accuracy of overdue cases is 52.1%, regular cases 93.7%, and the correctness in total 84.8%.
Based on the profit margins and competitions, how to raise the quality of credit and reduce the rate of overdue repayment becomes the key point which will be related to the successful operation of the financial industry in the future. Because of the high-rate of mortgage loans in consumer loans, the aims of this research are to investigate factors affecting the credit risk of applicants, and establish an assessment system for mortgages. It is hoped that the rate of overdue repayment can be brought down, that the quality of credit can be raised, and that a complete bank can be built.
Keywords: Mortgage loan, Logistic Regression, Risk Management, Credit Risk
|
|---|
