本文討論以下問題：
金融機構只參考個人特徵變量評估房屋貸款申請人是不是足夠的？我們發現，如果僅考量個人特徵變量，邏輯斯迴歸模型有0.78的機率正確預測貸款償付能力。然而，添加一個地區變量到模型後，機率提高至0.813.此外，如果我們消除迴歸中不顯著的統計變量後，機率更提高至0.847.結論得到地區變量，負債收入比和有無擔保人為影響貸款申請人償付能力的三個最重要的因素。最後，台北地區的貸款申請者比其他地區有5.749倍的機率發生逾期情形。

This thesis discusses the following question:
Is it enough for financial institutions to evaluate the mortgage applicants through the characteristic variables only? We find that, if only the characteristic variables are considered, a logistic regression model can have a probability of 0.78 to correctly predict the solvency of a loan. However, after adding a geographical variable into the model, the probability improves to 0.813. Moreover,if we eliminate the statistically insignificant variables from the regression, the probability can be improved furtherly to 0.847.Our results show that the geographical variable, debt-income ratio and having a guarantor or not are the three most important factors to influence on loan applicant’s solvency. Finally, a loan applicant has a 5.749 times higher probability to delinquency than on applicant from the other regions.

論文審定書 i
謝辭 ii
中文摘要 iii
英文摘要 iv
圖目錄 vii
第 一 章 緒論 1
第一節 研究背景 1
第二節 研究動機 4
第三節 研究目的 5
第 二 章 文獻回顧 6
第 三 章 模型與研究設計 10
第一節 Logistic Regression 信用評量模型 10
第二節 逾期放款的定義 12
第三節 銀行融資審核重要原則5P 13
第四節 資料來源 14
第五節 資料分類 15
第六節 信用評量與研究方法16
第 四 章 實證分析 17
第一節 解釋變數特徵交叉分析 17
第二節 探討不同LR的設定18
第三節 解釋變數的重要性 24
第 五 章 結論與建議 25
第一節 結論 25
第二節 建議 26
參考文獻 27

1. 王濟川、郭志剛（2003）,《Logistic 迴歸模型方法及應用》,五南出版社
2. 何太山 (1997) ,《運用區別分析建立商業放款信用評分制度》,政治大學企業管理研究所碩士論文
3. 李桐豪、呂美慧 (1990) , 「金融機構房貸客戶授信評量模式分析—Logistic迴歸之應用」,《金融財務季刊1-9》
4. 林建州（2000）,《銀行個人消費信用貸款授信風險評估模式之研究》,中山大學財務管理研究所碩士論文
5. 陳宗豪 (2000) , 《消費者小額信用貸款之信用風險研究 : 甄選的觀點，碩士論文》,國立中山大學人力資源管理研究所
6. 陳正昌、程炳林、陳新豐、劉子鍵 (1998) ,《多變量分析方法：統計軟體應用(4版) 》,五南出版社
7. 財團法人金融聯合徵信中心編(1995), 《金融機構授信管理要覽》
8. 馮志剛 (1996) ,《公營與民營金融機構個人擔保與信用放款授信評估之研究》,高雄工學院管科所碩士論文
9. 蔡士斌（2007）,《房屋貸款戶信用評量之研究 -以國內某一銀行為例》,中興大學應用經濟研究所碩士論文
10. 戴堅 （2004）,《個人消費性信用貸款授信評量模式之研究》,國立中正大學國際經濟研究所碩士論文
1. Basel Committee on Banking Supervision (1999), “Credit risk modeling: Current practices and applications, ” Basel Committee Publications, 8-54.
2. Brzezinski, J. R. and Knafl G. J.(1999), “Logistic Regression Modeling for Context-Based Classfication,” In: Proceedings tenth international workshop on database and expert systems applications, 755–759.
3. Capon, N. (1982), “Credit Scoring Systems: A Critical Analysis,” Journal of Marketing 82, Vol. 46 Issue 2, 82-90.
4. Espahibodi, P. (1991), “Identification of Problem Bank and Binary Choice Models,” Journal of Banking Finance, 15, 53-71.
5. Federal Reserve System Task Force on Internal Credit Risk Models (1988), “Credit risk models at major US banking institutions: Current state of the art and implications for assessments of capital adequacy,” Federal Reserve Bank Board of Governors, Supervisory Staff Reports, Washington, 1-50.
6. Gardner, M. J. and Mills, D. L. (1989), “Evaluating the Likelihood of Default on Delinquent Loans,” Financial Management , Vol. 18, No. 4, 55-62.
7. Harrell, F. E. and Lee, K. L. (1985), “A Comparison of the Discrimination of Discriminant Analysis and Logistic Regression under Multivariate Normality,” In P. K. Sen (Ed.): Biostatistics: Statistics in Biomedical, Public Health and Environmental Sciences. North-Holland: Elsevier Science Publishers, 333-343.
8. Hand, D. J. and Henley, W. E. (1997), “Statistical classification methods in consumer credit scoring: a review,” Journal of the Royal Statistical Society: Series A. Statistics in Society 160(3), 523–541.

9. Heinen, A., Kau, J., Keenan, D. C., Kim, M. L. and Slawson, C. (2011), “Contagion in Subprime Mortgage Defaults : a Composite Likelihood Approach,” Quantitative Management Initiative, 22-32.
10. Kim, J. S., Tokuhata, G. K. and Bratz, J. R. (1985), “Comparison of Multivariate Regression Analysis Between Logistic Model and the Least Square Model Using SAS Software,” SAS Users Group International Conference Proceedings, 1109-1112.
11. Lo, A. W. (1986), “Logit Versus Discriminant Analysis- A Specification Test and Application to Corporate Bankruptcies,” Journal of Econometrics, 31, 151-178.
12. Orgler, Y. E.(1970), “A Credit Scoring Model for Commercial Loans,” Journal of Money, Credit, and Banking, 435-445.
13. Steenackers, A. and Goovaerts, M. J. (1989), “A Credit Scoring Model for Personal Loans,” Insurance Mathematics Economics, 31-34.
14. Smalley, O. A. and Sturdivant, F. D. (1973), “The Credit Merchants: A History of Spiegel Inc,” Carbondale, IL: Southern Illinois University Press, 229.
15. Treacy, W. F. and Carey, M. (2000), “Credit risk rating at large US banks,” Journal of Banking and Finance, 24, 167–201.
16. Zhang, C. ( 2011), “Whether the Education Has Effect on Loan Defaults :A Theoretical and Empirical Study,” Industrial Engineering and Engineering Management (IE&EM), IEEE 18Th International Conference on ,Volume: Part 3, 1808-1810.
17. Zurada J. M., Foster B. P., Ward T. J. and Barker R. M. (1999), “Neural Networks Versus Logit Regression Model For Predicting Financial Distress Response Variables,” The Journal of Applied Business Research, Vol. 15, No 1, 21-29.