DC 欄位 值 語言 DC.contributor 財務金融學系 zh_TW DC.creator 陳侑宣 zh_TW DC.creator Yu-Hsuan Chen en_US dc.date.accessioned 2004-6-24T07:39:07Z dc.date.available 2004-6-24T07:39:07Z dc.date.issued 2004 dc.identifier.uri http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=91428027 dc.contributor.department 財務金融學系 zh_TW DC.description 國立中央大學 zh_TW DC.description National Central University en_US dc.description.abstract 為因應國際潮流及落實國內金融業風險管理理念，主管機關要求國內銀行在2006年底實施符合國際清算銀行標準之新版巴塞爾資本協定（New Basel Capital Accord）。因此，國內各家銀行莫不努力改善其風險管理體質，以達成主管機關的要求，同時亦使銀行的經營風險能得到有效的控管。 在台灣，商業銀行的企業放款絕大多數屬於信用風險難以有效掌控的中小企業。為協助銀行有效面對中小企業申貸案件，本文試圖建立一套中小企業的信用風險量化系統，讓銀行根據量化的信用風險值，隨時調整其授信政策。 本文是以樣本銀行提供之企業放款客戶作為研究對象，利用KMV模型與PFM®模型估計中小企業之預期違約機率，並參考Moody’s LossCalcTM方法，建立回收率模型，以及推算相關的違約損失率。然後，藉由上述的違約機率與損失率，和樣本銀行提供的其他資料，計算不同規模、產業…構面下，放款部位之信用風險值，俾藉以調整銀行的授信政策。 實證結果發現：（1）PFM模型適用於國內市場環境，且產業迴歸法優於中位數比對法；（2）產業、公司規模、有無擔保品、有無設定順位，以及客戶信用紀錄等因素顯著影響違約機率的高低；（3）回收率與設定比率、債權額度、經濟成長率、區域退票比率等變數呈現顯著的正向關係，而與失業率存在顯著的負向關係；（4）產業及公司規模能顯著影響未預期損失率（信用風險指標）；（5）借款客戶刻意藉由營業內外之操縱盈餘改變其申貸前三年之信用品質與風險。 zh_TW dc.description.abstract To meet supervisory requirement of Basel II and get efficient control on banking risks, commercial banks in Taiwan showed great efforts to improve their risk management system in recent years. This paper provides a credit risk quantification system for private firms, whose credit risks are most difficult to handle for commercial bankers. Also, commercial bankers can employ the structure of this paper to examine the performance of their loan policy. This paper employs Moody’s KMV model and Private Firm Model® to estimate the expected default frequency (EDF), and refers to Moody’s LossCalcTM to construct loss given default (LGD) forecasting model. Incorporating EDF, LGD with other given parameters, this paper calculates Credit VARs of different loan portfolios, and analyzes them across various dimensions. The empirical results show: (1) Private Firm Model is suitable for the banking market in Taiwan, and the industry regression approach dominates over comparables median approach. (2) Industry, firm size, collaterals, collateral seniority and credit records have significant effects upon probability of default. (3) The ratio of seniority setting, par value of loan, economic growth rate and regional bounced ratio all have significantly positive correlations with recovery rate, while the unemployment rate is negatively correlated with recovery rate. (4) Both industry and firm size have significant effects upon unexpected loss rate, which is the proxy of overall credit risk in this paper. (5) Earning management behavior has a significant effect upon credit risk. en_US DC.subject 違約損失率 zh_TW DC.subject 信用風險值 zh_TW DC.subject 盈餘操縱 zh_TW DC.subject 授信政策 zh_TW DC.subject LossCalc zh_TW DC.subject PFM zh_TW DC.subject 違約機率 zh_TW DC.subject 回收率 zh_TW DC.subject KMV zh_TW DC.subject KMV en_US DC.subject Private Firm Model en_US DC.subject LossCalc en_US DC.subject probability of default en_US DC.subject recovery rate en_US DC.subject loss given default en_US DC.subject credit VAR en_US DC.subject earning management en_US DC.subject loan policy en_US DC.title 商業銀行如何利用信用風險值檢視授信政策 zh_TW dc.language.iso zh-TW zh-TW DC.title How to Use Credit VAR to Examine Commercial Banks' Loan Policy en_US DC.type 博碩士論文 zh_TW DC.type thesis en_US DC.publisher National Central University en_US