碩士 / 國立高雄應用科技大學 / 金融系金融資訊碩士在職專班 / 103 / This paper is to investigate the factors of credit risks of the applicants in mortgage loan on default. We build up the credit risk model with twelve risk factors. Cox regression model and logistic regression model are applied to analyze the correlations between risk factors and default.
The empirical results show that the four variables -Occupation, Annual income, Guarantors, and the type of house, are significantly negatively related with default. In addition, Cox regression model shows that the curve of cumulative survival rate function, the curve of one minus cumulative survival rate and the curve of cumulative hazard rate function are steeper between 12 and 20 months after approval. It means that the debtors tend to be delinquent in this period and then the curve will tend to be flat. Therefore, Bank should pay more attention to the mortgages between one and two years after approval.
The logistic regression model gets the same results with Cox regression model. Overall, the ability of prediction of logistic regression model is reasonable.
| Identifer | oai:union.ndltd.org:TW/103KUAS1213028 |
| Date | January 2015 |
| Creators | Ya-Chin, Lin, 林雅琴 |
| Contributors | Kun-Min, Hsieh, 謝坤民 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 93 |
Page generated in 0.0119 seconds