臺灣博碩士論文加值系統

全球金融體系連結性高，一旦發生類似於2007年的美國次貸危機與2010年的歐洲債務危機必定會掀起全球經濟環境的一陣波瀾，可能會造成國內經濟遭受打擊而增加金融機構之逾期放款金額。因此，許多金融機構開始建構一套客觀且公正的風險評估模型，並藉此判斷借款客戶的信用好壞以預防壞帳或呆帳之發生。大部分金融機構在建構風險評估模型時，變數僅考慮借款企業的資歷、財力與背景等內部資訊。考量到總體經濟與景氣循環可能也會影響企業違約風險，過去有文獻將總體經濟指標加入模型內。然而，過去的研究大多是以主觀的方式選取要加入模型的總體經濟指標，本研究則是透過皮爾森相關分析與主成分分析進行篩選，將篩選後的總體經濟指標加入模型內成為新增變數，並且應用兩階段整體學習方法整合三種分類器（邏輯斯迴歸、支持向量機與梯度提升決策樹）建構模型。利用台灣某金融機構提供的實際中小企業借款資料作為驗證資料，並使用本研究所設計之應用整體學習結合總體經濟指標所建構的風險評估模型進行預測，得到AUC值0.7996之結果，優於其他常見的單階段分類器模型；加入總體經濟指標也確實能提升模型的預測能力。

Due to the high connection of the global financial system, the international financial crisis may have a significant influence on the domestic economy and increase the number of non-performing loans from financial institutions. As a result, many financial institutions have begun to construct an objective and fair risk assessment model. However, most financial institutions only take internal information about borrowing SMEs into account when constructing the model. Therefore, considering the macroeconomic environment may affect the risk of default, this thesis selects macroeconomic indices through Pearson Correlation Analysis and Principal Component Analysis to become new variables. On the other hand, the two-stage ensemble learning method, which integrates three classifiers (Logistic Regression, Support Vector Machine, and Gradient Boosting Decision Tree) is applied to construct the model in the thesis. A financial institution in Taiwan provides the actual SMEs loan data as the verification data. According to the result, the risk assessment model proposed in this thesis outperforms other common single-stage classifier models. Furthermore, adding the macroeconomic indices in the model is also proved to enhance the prediction performance.

摘要 i
ABSTRACT ii
目錄 iv
圖目錄 vi
表目錄 vii
第一章 緒論 1
1.1研究背景與動機 1
1.2研究目的 3
1.3研究架構 3
第二章 文獻探討 5
2.1信用評等與風險評估簡介 5
2.1.1信用評等與風險評估定義 5
2.1.2新巴塞爾資本協定與內部評等法 6
2.1.3金融機構授信原則 7
2.2授信風險評估模型 8
2.3將經濟因子納入授信風險評估模型之研究 12
2.4模型的基礎理論 12
2.4.1皮爾森相關分析 13
2.4.2主成分分析 13
2.4.3邏輯斯迴歸 14
2.4.4支持向量機 16
2.4.5梯度提升決策樹 18
2.4.6整體學習 19
2.5 ROC曲線與AUC 21
第三章 研究方法 24
3.1研究架構 24
3.2建構整體學習結合總體經濟指標之風險評估模型 25
第四章 實例驗證 30
4.1建構風險評估模型 30
4.2驗證本研究所推薦之風險評估模型 42
第五章 結論與建議 44
5.1研究貢獻 44
5.2未來研究與建議 44
附錄一 45
參考文獻 46

王進（2008）‧信用管理基礎教程‧中國：中國金融出版社。
林佳蓉（2008）‧信用風險模型之發展與衡量-以中長期資金運用制度為例‧高雄市：國立中山大學財務管理學系碩士論文。
林金龍、彭俊能（2012）‧臺灣總體經濟因子對信用風險的影響─專案研究計畫成果與檢討‧人文與社會科學簡訊，14（1），78-86。
沈大白、張大成、劉宛鑫（2002）‧信用評等模型之簡介‧中國商銀月刊，21（11），1-5。
吳莉安（2005）‧中小企業違約信用風險評估流程‧新竹市：國立交通大學工業工程與管理學系碩士論文
吳冠誌（2018）‧應用極端梯度提升決策樹建構金融機構信用風險評估模型‧新竹市：國立交通大學工業工程與管理學系碩士論文
游日傑（2004）‧考慮違約相關性下，以「信用價差違約模型」評價信用衍生性商品‧桃園市：國立中央大學財務金融學系碩士論文。
黃博怡、張大成、江欣怡（2006）‧考慮總體經濟因素之企業危機預警模型‧金融風險管理季刊，2（2），75-89。
農業金融局（無日期）‧授信實務‧取自https://www.boaf.gov.tw/site/boaf/public/Attachment/992314201471.doc。
楊蓁海（2005）‧新版巴賽爾資本協定與銀行信用風險測度模型的發展：兼論對我國銀行體系與央行政策的影響‧中央銀行季刊，27（1），47-86。
褚鴻烜（2012）‧建立以決策樹為基礎之短期違約放款案件信用風險評估模型‧新竹市：國立交通大學工業工程與管理學系碩士論文
歐明鴻（2017）‧利用極端梯度提升決策樹建構液晶面板自動光學檢測瑕疵分類模型‧新竹市：國立交通大學工業工程與管理學系碩士論文
Altman, E. I. (1968). Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. The Journal of Finance, 23(4), 589-609.
Baesens, B., Setiono, R., Mues, C., & Vanthienen, J. (2003). Using neural network rule extraction and decision tables for credit-risk evaluation. Management Science, 49(3), 312-329.
Basel Committee on Banking Supervision. (2004). International Convergence of Capital Measurement and Capital Standards. Retrieved from: https://www.bis.org/publ/bcbs107.pdf
Beaver, W. H. (1966). Financial ratios as predictors of failure. Journal of Accounting Research, 71-111.
Berkson, J. (1944). Application of the logistic function to bio-assay. Journal of the American Statistical Association, 39(227), 357-365.
Breiman, L. (1996). Bagging predictors. Machine Learning, 24(2), 123-140.
Cattell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioral Research, 1(2), 245-276.
Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785-794). ACM.
Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3), 273-297.
Desai, V. S., Crook, J. N., & Overstreet Jr, G. A. (1996). A comparison of neural networks and linear scoring models in the credit union environment. European Journal of Operational Research, 95(1), 24-37.
Dietterich, T. G. (2000). Ensemble methods in machine learning. In International workshop on multiple classifier systems. (pp. 1-15). Springer, Berlin, Heidelberg.
Dutta, A., Bandopadhyay, G., & Sengupta, S. (2015). Prediction of stock performance in indian stock market using logistic regression. International Journal of Business and Information, 7(1).
Dutta, S., & Shekhar, S. (1988). Bond rating: a non-conservative application of neural networks. In IEEE Int Conf on Neural Networks. Publ by IEEE.
Edmister, R. O. (1972). An empirical test of financial ratio analysis for small business failure prediction. Journal of Financial and Quantitative Analysis, 7(2), 1477-1493.
Fong, Y., Yin, S., & Huang, Y. (2016). Combining biomarkers linearly and nonlinearly for classification using the area under the ROC curve. Statistics in Medicine, 35(21), 3792-3809.
Friedman, J. H. (2001). Greedy function approximation: a gradient boosting machine. Annals of Statistics, 1189-1232.
Haldeman, R. G., & Narayanan, P. (1977). Zeta Analysis: A New Model to Identify Bankruptcy Risk of Corporations. Journal of Banking and Finance, 1, 29-54.
Hajian-Tilaki, K. (2013). Receiver operating characteristic (ROC) curve analysis for medical diagnostic test evaluation. Caspian Journal of Internal Medicine, 4(2), 627.
Han, G. S., Yu, Z. G., & Anh, V. (2014). A two-stage SVM method to predict membrane protein types by incorporating amino acid classifications and physicochemical properties into a general form of Chou's PseAAC. Journal of Theoretical Biology, 344, 31-39.
Hanley, J. A., & McNeil, B. J. (1982). The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology, 143(1), 29-36.
Huang, J. J., Tzeng, G. H., & Ong, C. S. (2006). Two-stage genetic programming (2SGP) for the credit scoring model. Applied Mathematics and Computation, 174(2), 1039-1053.
Huang, C. L., Chen, M. C., & Wang, C. J. (2007). Credit scoring with a data mining approach based on support vector machines. Expert Systems with Applications, 33(4), 847-856.
Koutanaei, F. N., Sajedi, H., & Khanbabaei, M. (2015). A hybrid data mining model of feature selection algorithms and ensemble learning classifiers for credit scoring. Journal of Retailing and Consumer Services, 27, 11-23.
Le, Q. K., Nguyen, H. K. K., Huynh, Q. H., & Huynh, Q. L. (2017). Analyzing Sleep Microstructure by Using Support Vector Machine. In International Conference on the Development of Biomedical Engineering in Vietnam (pp. 307-312). Springer, Singapore.
Lee, T. S., Chiu, C. C., Lu, C. J., & Chen, I. F. (2002). Credit scoring using the hybrid neural discriminant technique. Expert Systems with Applications, 23(3), 245-254.
Lin, S. L. (2009). A new two-stage hybrid approach of credit risk in banking industry. Expert Systems with Applications, 36(4), 8333-8341.
Liu, W., Zhang, M., Luo, Z., & Cai, Y. (2017). An ensemble deep learning method for vehicle type classification on visual traffic surveillance sensors. IEEE Access, 5, 24417-24425.
Luo, S., Cheng, J., & Ao, H. (2015). Application of LCD-SVD technique and CRO-SVM method to fault diagnosis for roller bearing. Shock and Vibration.
Mester, L. J. (1997). What’s the point of credit scoring?. Business Review, 3, 3-16.
Odom, M. D., & Sharda, R. (1990). A neural network model for bankruptcy prediction. In Neural Networks, 1990., 1990 IJCNN International Joint Conference on (pp. 163-168). IEEE.
Ohlson, J. A. (1980). Financial ratios and the probabilistic prediction of bankruptcy. Journal of Accounting Research, 109-131.
Opitz, D. W., & Maclin, R. (1999). Popular ensemble methods: An empirical study. J. Artif. Intell. Res.(JAIR), 11, 169-198.
Orgler, Y. E. (1970). A credit scoring model for commercial loans. Journal of Money, Credit and Banking, 2(4), 435-445.
Pearson, K. (1895). Note on regression and inheritance in the case of two parents. Proceedings of the Royal Society of London, 58, 240-242.
Pearson, K. (1901). LIII. On lines and planes of closest fit to systems of points in space. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 2(11), 559-572.
Peng, R. (2017). Personal Credit Assessment Model Based on Stacking Ensemble Learning Algorithm.
Quinlan, J. R. (1986). Induction of decision trees. Machine Learning, 1(1), 81-106.
Shahabi, H., Khezri, S., Ahmad, B. B., & Hashim, M. (2014). Landslide susceptibility mapping at central Zab basin, Iran: a comparison between analytical hierarchy process, frequency ratio and logistic regression models. Catena, 115, 55-70.
Shin, K. S., Lee, T. S., & Kim, H. J. (2005). An application of support vector machines in bankruptcy prediction model. Expert Systems with Applications, 28(1), 127-135.
Thyagharajan, A., & Routray, A. (2017, July). An ensemble metric learning scheme for face recognition. In Multimedia and Expo (ICME), 2017 IEEE International Conference on (pp. 115-120). IEEE.
Valiant, L. (2013). Probably Approximately Correct: NatureÕs Algorithms for Learning and Prospering in a Complex World. Basic Books (AZ).
WAN, P., Wu, C., Lin, Y., & Ma, X. (2016). A recognition model of driving anger based on physiological features by ROC curve analysis. The 95th Annu Meeting of Transp Res Board , 10-14.
Wang, G., & Ma, J. (2012). A hybrid ensemble approach for enterprise credit risk assessment based on support vector machine. Expert Systems with Applications, 39(5), 5325-5331.
Wolpert, D. H. (1992). Stacked generalization. Neural Networks, 5(2), 241-259.
Zhang, Y., & Wang, G. (2015). Tool condition monitoring technology for CFRP drilling based on heterogeneous ensemble learning model. Aerospace Materials & Technology, 6, 008.
Zhou, J., & Bai, T. (2008). Credit risk assessment using rough set theory and GA-based SVM. In Grid and Pervasive Computing Workshops, 2008. GPC Workshops' 08. The 3rd International Conference on (pp. 320-325). IEEE.