[[alternative]]Behavioral scoring model for bank customers using data envelopment analysis

[[abstract]]行為評分模式(behaivoral scoring)可以讓企業評估既有顧客的信用風險與消費狀況，進而衡量顧客貢獻度，對金融業而言是一個相當重要的風險管理工具。然而現存之行為評分模式，其分析結果多半只將分析對象區分為好顧客與壞顧客，缺乏針對分析結果提供具體改善方向的能力。資料包絡分析法 (data envelioment analysis, DEA) 是一個無母數的多目標決策工具，用以衡量決策制定單位 (decision making unit, DMU) 的相對效率，其特點在於提供無效率之決策制定單位明確的效率改善方向，因此廣泛用於各項領域的績效評估議題上。本研究應用資料包絡分析法建構銀行顧客的行為評分模式，並利用某銀行所提供之信用卡資料進行實證。所提方法透過DEA將信用卡持卡人區分出貢獻度高與貢獻度低之兩種類型顧客，之後再針對低貢獻度的顧客，利用DEA的差額分析結果，提供銀行改善的方向，以將低貢獻的顧客轉化成高貢獻的顧客。實證結果顯示，所提方法能有效的將持卡人區分為高貢獻與低貢獻客戶，並且能提供銀行明確的方向，將低貢獻的顧客轉化成高貢獻的顧客，達成個人化行銷或客製化管理的目的。[[abstract]]Behavior scoring is an important part of risk management in financial institutions, which is used to help banks make better decisions in managing existing customers by forecasting their future credit risk and expenditure performance. The existing behavior scoring methods usually generate the results of “good creditor” or “bad creditor” from customers, however, they are lack of improving abilities for classification results. This study proposes a behavior scoring model based on data envelopment analysis (DEA) to manage existing credit card customers in a bank. DEA is a method of measuring the relative efficiencies of decision making units (DMUs). The most important advantage of DEA is providing an indeed improvement for decision making unit (DMU). The proposed method uses DEA model to classify the customers into high contribution customers and low contribution customers. Then, the low contribution customers will be examined by using the slack analysis of DEA model to promote their contributions. A real credit cardholder dataset in a Taiwan commercial bank is selected as the experimental data to demonstrate the performance of the proposed method. The experiment results showed that the proposed method can provide indeed directions for bank to improve the contribution of the low contribution customers, and facilitates marketing strategy development.[[journaltype]]國內[[ispeerreviewed]]Y[[booktype]]紙本[[countrycodes]]TW