碩士 / 輔仁大學 / 應用統計學研究所 / 96 / The objective of this study is to build a credit scoring system to guard against management risk for the credit-issue financial corporations. A pragmatic devise merging the decision tree CHAID techniques with logistic regression is proposed and applied to the model building. This twofold approach uses the classification tree algorithm to generate classifiers from the bank credit card data and the logistic regression model to estimate the relative contribution of predictor variables to the classification rule. It can effectively identify the most predictive subsets and estimate the probability of any credit applicant being a good risk given the characteristics vector.
| Identifer | oai:union.ndltd.org:TW/096FJU00506016 |
| Date | January 2008 |
| Creators | Hsu Jung-Chieh, 許榮傑 |
| Contributors | Rwei-Ju CHUANG, 莊瑞珠 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 127 |
Page generated in 0.1067 seconds