- 學位論文
汽車貸款風險預測之灰色決策分析 —以C融資公司為例
Using Grey Decision Making to Forecast Automobile Loan Risk -A Case of C Financing Company
摘要
自1911年政府開放新銀行設立後，多數金融業者因辦理企業貸款陸續發生掏空及惡性倒閉，轉而發展消金業務為主力。由於金融機構消費性貸款商品不斷地推陳出新，各家金融業者為達市場占有率，紛紛提高客戶額度、降低放款利率及鬆綁基本放款條件，直到2005年雙卡風暴侵襲國內，造成消費大眾之信用卡及現金卡使用過度氾濫，也因此由銀行公會出面訂定「債務協商機制」，與卡債族協商還款。許多銀行也因此提列大量備抵呆帳，獲利轉盈為虧。本研究動機及目的為制定新的汽車貸款評估模式，即早發現影響逾期還款及呆帳發生之關鍵因素，以達到更精準的判斷客戶風險，並可兼顧業務發展及授信品質，對金融業者發展汽車貸款能有極大助益。 本研究以德爾菲專家問卷法，依據過去學者所彙整汽車貸款風險因子，以確立本研究之關鍵變數，再以C融資公司汽車貸款客戶資料為測試資料，應用灰關聯分析法建構汽車貸款風險預測模型，經由68筆測試案例並以WEKA(3.8.2版)分析得知，本研究以灰關聯分析法應用在預測汽車貸款風險上，確實較能優於其他數據分析法(神經網絡、支援向量機、決策樹、多項羅吉特)。而對個案公司而言，為因應目前汽車貸款低利率環境，能減少更多的呆帳損失，就能為企業帶來更多的獲利，本研究使用灰關聯分析法，依其特異度(Specificity)對汽車貸款產業之降低損失及實質獲利相對重要。 研究結論在汽車貸款業務之審查、業務方面皆有顯著貢獻，在審查方面可簡化現有審查流程且快速判斷風險，在業務方面可使第一線業務人員能實際掌握客戶履約之關鍵因素，更能迅速、準確且風險無虞下拓展汽車貸款業務，達成企業獲利模式。
並列摘要
Since 1911, when the government deregulates the commercial bank entry barriers, most of the finance companies become tunneling and fraudulent bankrupt one after another, then turn to the consumer finance. Every finance company tries to reach a higher market share since the financial institution keeps rolling out the new consumer loan products. In order to reach this goal, they started to increase their customer's line of credit, decline the prime rate, and lose the condition of lending. Until 2005, the dual-SIM storm hit the domestic market, it causes that excessive spread of credit cards and cash cards in general public. Based on this situation, the Bankers Association of the Republic of China drawn up the "Debt Negotiation Mechanism" and negotiate repayment for credit card debtors. Many banks provisioned loan loss, profits turn gain into a loss. The purpose of this study was to draw up a new evaluation model. It can early detection the key factors of overdue repayment, and avoid loan loss. Also, balance the business development and the lending quality, and contribute to financing companies to develop the automobile loans. This research use Delphi Method. Based on the previous related literature, this research collected automobile loan's risk factor to confirm the key variables. Then, use C financing company's automobile loan customer's data as the testing data, and applies Gray Relation Analysis to build up the automobile loan risk prediction model. Then, 68 testing case analyzed with the WEKA(3.8.2 Edition). This research surely confirms that use Gray Relation Analysis in predicting the risk of automobile loan is better than other data analysis methods (neural networks, support vector machine, decision tree, Multinomial Logit). For the case company, react to the current low interest rate of the automobile loan. In this situation, can reduce the bad debts is same as bring a lot of profits to the company. The research contributes significantly to automobile loan's due diligence. In the aspect of due diligence, this research can simplify the process, and make the risk judgment quickly. For the business, this research can help the frontline employee keep abreast of the key factors of customer performance, and expand the business scope no only fast and accurate but also without the risk. As mentioned above, this research finds out a model of corporate profit.
參考文獻
延伸閱讀
- 謝忠榮（2004）。授信風險影響因素之探討-以汽車貸款為例〔碩士論文，亞洲大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0118-0807200916281432
- 陳明輝（2009）。從次貸風暴探討對國內金融資產證券化市場之影響並以C公司為例〔碩士論文，國立清華大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0016-1111200916011102
- 黃佐民（2010）。房屋貸款授信風險評估研究－以ｃ銀行為例〔碩士論文，中原大學〕。華藝線上圖書館。https://doi.org/10.6840/cycu201000251
- Lin, M. C. (2004). 消費性貸款授信風險評估之研究—以X銀行為例 [master's thesis, Tatung University]. Airiti Library. https://www.airitilibrary.com/Article/Detail?DocID=U0081-0607200917232362
- 黃昱珳（2022）。The study of Early Warning Model of Corporate Credit Risk〔碩士論文，淡江大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0002-0207202217482400