邏輯斯迴歸模型運用在女性信用卡評分制度之研究

Study of the Female Credit Card Scoring via the Modeling of Logistic Regression and Tree Structure

10.29698/FJMR.200701.0006

莊瑞珠(Rwei-Ju Chuang)

信用風險 ； 評分卡 ； 邏輯斯迴歸模型 ； credit risk ； credit scoring ； logistic regression model

輔仁管理評論

14卷1期（2007 / 01 / 01）

127 - 154

繁體中文

本研究以預測信用風險的方向，評估最能衡量個人信用、償債能力的預測變數，茲依照各個因素對於個人信用狀況的影響程度給予不同的權重，做出事前的風險量化研究，策略上評定是否發給個案信用卡，以期提高信用良好顧客比例，減少銀行呆帳的發生。首先依關聯性分析，發掘與顧客信用好壞有關的決定因素。進而透過這些因素，利用勝算比觀念，決定迴歸估計係數的權重，依此建立女性信用卡持有人之信用風險較完整評估標準。本研究顯示分別以教育程度、有無不動產、職位、行業為顧客信用好壞的顯著相關因素。在線性條件預測模型下，建立多變量邏輯斯分析模型，並依此結果製成業界所能直接應用的評分卡，揭露依評分模型所計算出評分分數等級與所屬百分位點，協助資訊使用者瞭解評分所代表的實際風險意義。並評估風險模型適用性與區分能力診斷，並偵測發卡銀行女性顧客群的信用組合，作為建構以個人信用評分分級為基礎的信用卡差別利率授信審核政策之參考。

The volume of credit business in the female group has greatly expanded and the use of credit scoring through the evaluation of large credit portfolio becomes crucial to guard against any management risk. The objective of this study is to devise a credit scoring system for credit granting decisions. We describe statistical method to create scorecards and show how the result of the model is applied to calculate score point weights. Scorecards are built using the logistic regression method which estimates the relationship between the individual characteristics and the log of the odds (risk) so that the score point weights can be calculated directly from the regression coefficients. The model performance is usually monitored by the model validation and classification error. We propose an alternative measure for the better power of model discriminations and the direct use of credit-granting decisions.

1. Boyle, M.,Crook, J.N.,Hamilton, R.,Thomas L.C.,Thomas (ed.)(1992).Proceedings of the IMA Conference on Credit Scoring and Credit Control.Oxford:Crook and Edelman, Clarendon Press.
2. Capon, N.(1982).Credit Scoring Systems: A Critical Analysis.Journal of Marketing,46,82-91.
3. Carter, C.,Catlett, J.(1987).Assessing Credit Card Applications Using Machine Learning.EEE Expert,2,71-79.
4. Davis, R.H.,Edelman, D.B.,Gammerman, A.J.(1992).Machine-learning Algorithms for Credit-Card Applications.IMAJ. Mathematics applied in Business and Industry,4,43-52.
5. Eisenbeis, R.A.(1978).Problems in Applying Discriminant Analysis in Credit Scoring Models.Journal of Banking and Finance,2,875-900.
6. Gilbert, L.R.,Menon, K.,Schwartz, K.B.(1990).Predicting Bankruptcy for Firms in Financial Distress.Journal of Business Finance Accountings,17(1),161-171.
7. Grablowsky, B.J.,Kalley, W.K.(1981).Probit and Discriminant Functions for Classifying Credit Applicants; A Comparison.Journal of Economics and Business,33,254-261.
8. Henley, W.E.,Hand, D.J.(1996).A k-NN Classifier for Assessing Consumer Credit Risk.The Statistician,65,77-95.
9. Makowski, P.(1985).Credit Scoring Branches out.The Credit World,75,30-37.
10. Radding, A.(1992).Credit Scorings: New Frontier.Bank Management,68,57-62.
11. Royston, P.,Altman, D. G.(1994).Regression Using Fractional Polynomials of Continuous Covariates: Parsimonious Parametric Modeling (with Discussion).Applied Statistics,43,429-467.
12. Srinivasn, V.,Kim, Y.H.(1987).Credit Granting: A Comparative Analysis of Classification Procedures.Journal of Finance,42,665-683.
13. Steenackers, A.,Goovaerts, M.J.(1989).A Credit Scoring Model for Personal Loans.Insurance Mathematics Economics,8
14. Yobas, M.B.,Crook, J.N.,Ross, P.(1997).Working Paper 97/2, Credit Research Center, University of Edinburgh.

1. 陳穆貞、莊瑞珠(2006)。金融機構住宅房屋貸款信用評分系統之建構研究。住宅學報，15(2)，65-90。
2. 葉建良、梁德馨(2008)。消費者信用貸款違約風險評估之研究—以CART分類與迴歸樹建模。中山管理評論，16(3)，465-506。