Who Are Creditworthy? A Two-Stage Behavioral Scoring Model Identifies

整合類神經網路及資料包絡分析法建構銀行顧客之行為評等模式

陳怡妃(I-Fei Chen) ； 李永新(Yung-Hsin Lee)

《Journal of Data Analysis》 7卷6期 (2012/12) Pp. 1-28

https://doi.org/10.6338/JDA.201212_7(6).0001

摘要

有鑑於金融業對風險管理日益重視，且授信決策須同時考量顧客未來之信用風險及利潤貢獻程度；因此本研究試圖以二階段方法建構信用卡顧客行為評等模式。首先分別利用類神經網路（ANNs）及卡方自動交互探測法（CHAID）等分類器預測顧客未來還款行為，而於第二階段根據分類結果，進一步援引資料包絡分析法（DEA），探討不同還款狀況下個人績效之相對效率，並藉此效率以驗證前一階段之分類結果，能有效降低對顧客未來還款狀態之誤判，同時亦可透過DEA所得之差額分析，提供銀行授信管理方向與建議，將低貢獻之客轉換為高貢獻之顧客，致使金融業者風險最小利潤最大化。

關鍵字

類神經網路 ； 資料包絡分析（DEA） ； 行為評等 ； 資料探勘 ； 卡方自動交互探測法（CHAID）

並列摘要

Since the databases banks use for analysis of cardholders' behaviors are increasingly sophisticated and classification techniques today rarely offer 100% classification accuracy so as to possibly incur a considerable loss associated with type II errors, the prediction of cardholders' future payment behaviors has been referred to as a difficult task in the credit industry.An empirical study validates that the two-stage cardholder behavioral scoring model proposed by this paper, with merits of highly-performed classifiers and data envelopment analysis (DEA), not only enables banks to identify the creditworthy cardholders who are valuable without misclassification risks but also provides guidelines to improve contributions of each inefficient cardholder for card issuer profitability.

並列關鍵字

artificial neural networks (ANNs) ； data envelopment analysis ； behavioral scoring ； data mining ； Chi-square automatic interaction detector

參考文獻

Baesens, B.,Van Gestel, T.,Stepanova, M.,Van den Poel, D.,Vanthienen, J.(2005).Neural network survival analysis for personal loan data.Journal of Operations Research Society.56(9),1089-1098.

Banasiak, M.,O'Hare, E.(2001).Behavior scoring.Business Credit.103(3),52-55.

Banker, R.D.,Chang, H.,Lee, S.Y.(2010).Differential impact of Korean banking system reforms on bank productivity.Journal of Banking & Finance.34(7),1450-1460.

Berry, M.J.,Linoff, G.(1997).Data mining techniques for marketing, sales and customer support.Indianapolis:Wiley.

Berry, M.J.A.,Linoff, G.(2004).Data mining techniques.Indianapolis:Wiley.