企業投資資產的資金來源不外乎來自權益的自籌或是來自負債的舉借，其資金是否有效投資及配置於生財資產，即成為企業創造營收及獲利的關鍵，而台灣企業之資本能力相對國外較弱，高固定資產公司多以銀行借款充當第二股本，且銀行授信往往需徵提擔保品，未發展出信用有價概念，使企業可能將資金分配於低效率不動產上。本研究係以債權人角度，針對高固定資產公司之財務數據，利用杜邦分析法及資料包絡分析法(DEA)，建立一套關鍵有效之財務控管指標，用以預測財務危機發生之可能性，做為銀行業授信判斷之初步參考。

本研究主要關鍵要點有下列五項：一、將現金流量指標及負債風險觀點帶入杜邦分析法之評估範圍。二、拆解杜邦分析比率分為經營能力不佳及體質安全不佳兩個構面共5項指標，做為判別高固定資產公司財務危機之關鍵數據。三、將該關鍵數據依其與財務危機公司之關聯性，分為3個投入項及2個產出項做為資料包絡分析法(DEA)之變數。四、修改資料包絡分析法(DEA)之求取效率前緣模型，改為反向模型以求取不效率前緣，用來聚焦於財務危機公司，以利直接做判別分析。五、透過求取之不效率前緣公司，驗證實際發生財務危機之公司，並以該財務比率做為授信控管門檻。

本研究以台灣經濟新報(TEJ)資料庫之2006年至2007年跨期之上市櫃電子科技業財務數據為主，並用以預測2008年發生財務危機之公司，其中，高固定資產公司係以固定資產佔總資產50%以上且以固定資產前50大之公司為樣本，另排除海外掛牌、信用極佳或特差者。研究顯示，6家實際違約公司，以前一年及前二年財務均數預測，均能被反向資料包絡分析法(DEA)命中，判別率達100%，但也發生1至2家型一誤差，意即將非危機公司列為危機公司，惟因銀行授信保守立場，該型一誤差影響遠較型二誤差低，且據此結果所形成之違約財務比率臨界值，尚能成為預測後續財務危機公司之參考。

The funds that enterprises invest in assets come from either equity financing or debt financing. Whether these funds are effectively invested in the money-making assets is the critical issue of the creation of revenues in enterprises. Whereas, since the Taiwanese enterprises’ capital capacity are relatively lower than the foreign ones, they mostly take bank loans as a second equity; and since the concept of credit-worthiness is not yet developed, the collaterals are usually required to apply for a bank loan. Therefore, the enterprises may deploy the funds in low-efficiency real estate. From the creditors’ point of view, focusing on high capex companies, the present study makes use of DuPont Analysis and Data Envelopment Analysis (DEA) to build a set of critical and effective financial control indicators in order to predict the risk of financial crises and thus provide them for banks as a preliminary reference at the moment of the assessment of credit granting.

The present study contains five key points, as follows: (1) EBITDA and the concept of liability risk are taken into the scope of assessment of DuPont Analysis. (2) DuPont Analysis ratio is divided into two dimensions, i.e. poor profitability and poor constitution security, including 5 indicators, which are to be used as the critical data to assess the financial crises of the high capex enterprises. (3) According to the correlation between the critical data and enterprises facing a financial crisis, the critical data is divided into 3 input variables and 2 output variables as the variables in DEA. (4) The model for obtaining efficient frontier in DEA is modified as reverse model in order to obtain the inefficient frontier and then focus on the enterprises facing a financial crisis in order to facilitate the direct assessment and analysis. (5) Through the enterprises reaching inefficient frontier, the enterprises actually facing a financial crisis are verified and the financial ratio is taken as the controlling threshold of credit granting.

According to the financial data of the listed companies in the electronic technology industry from 2006 to 2007 from Taiwan Economic Journal (TEJ) database, the present study predicts the enterprises that will face a financial crisis in 2008. For this, the high capex company are defined as the enterprises whose fixed assets account for more than 50% of their total assets and whose fixed assets are ranked as top 50, excluding the overseas listed enterprises or the enterprises having extremely good or extremely poor credit scores. The present study shows that 6 enterprises actually breaching financial covenants can be predicted using their financial mean one or two years ago and the reverse DEA, representing 100% discriminatory ability. However, a Type-I error is also found in one or two enterprises, that is, the non-distressed enterprises are classified as distressed enterprises. But, owing to the banks’ conservative stance of credit granting, such Type-I error has a smaller effect than a Type-II error and the critical value of breach in financial ratio accordingly obtained can still become a reference for the prediction of the subsequent enterprises facing a financial crisis.

中文部分：
1.王冠倩，2004，「金融機構放款效率之評估-資料包絡分析法之應用」，國立高雄第一科技大學金融營運系，碩士論文。
2.白欽元，2003，「國內中小企業財務危機預警模型之研究」，國立交通大學經營管理研究所，碩士論文。
3.台灣經濟新報編輯部，2009/7，「信用評等專題」，貨幣觀測與信用評等2009年7月份P.109-110。
4.台灣經濟新報編輯部，2010/1，「信用評等專題之二」，貨幣觀測與信用評等2010年1月份P.120。
5.谷月涵，2005/10/24，「日本金改之路」，今周刊P.16，台北：金周文化事業。
6.李冠皓、陳惠玲，2012/3，「TCRI方法論--違約狀態之解除規則」，台灣經濟新報，貨幣觀測與信用評等2012年3月。
7.林哲群，2011，清華科管院MBA財務管理課程講義。
8.林淑萍、黃紹彥、蔡昆霖，2007/6，「企業危機預警模式之研究-DEA-DA、邏輯斯迴歸與類神經網路之應用」，會計與公司治理，第四券第一期。
9.高強、黃旭男，2011/9，「管理績效評估--資料包絡分析法」，初版，台北：華泰文化。
10.帥嘉珍，2005，「結合約略集合理論與分層資料包絡分析之財務預警模式」，國立交通大學資訊管理所，博士論文。
11.郭朝文、李世偉、林娟憶，2007，「使用DEA分析選擇股票投資標的-以IC設計產業為例」，清華大學工業工程管理學系，2007決策分析研討會。
12.張漢傑，2009，「企業現金流量的虛與實」、「解讀淨值報酬率的密碼-杜邦財務比率」，會計研究月刊，278期P.70-77及279期P.86-93，上海財經大學MBA學院客座教授。
13.張漢傑，2010/8，清華科管院MBA會計課程，專題演講「兩岸危機公司財報預警實務分析」，上海財經大學MBA學院客座教授。
14.鄭家年，2011/12，清華科管院MBA決策分析課程，專題演講「確定性決策分析模式」課程講義。
15.潘玉葉，1990，「台灣上市公司財務危機預警分析」，私立淡江大學管理科學研究所，博士論文。
16.賴季柔，2000，「企業失敗危機的預測-現金管理模式、財務比率模式、現金流量模式與混合模式的比較」，私立輔仁大學管理學研究所，碩士論文。
17.簡禎富，2011/9，「決策分析與管理」，修訂版，台北：雙葉書廊。
18.簡禎富，2011/11，清華科管院MBA決策分析課程，「資料包絡分析法」講義。
19.魏曉琴，2004，「財務危機預警模型之研究-以台灣地區上市公司為例」，國立交通大學財務金融研究所，碩士論文。

外文部分：
1.Altman, E. I., 1968, Financial ratios: Discriminate analysis and the prediction of corporate bankruptcy. Journal of Finance 23(4):589-609.
2.Altman, E. I., R. G. Hadelman, and P. Narayanan, 1977, ZETA analysis: A new model to identify bankruptcy risk of corporations. Journal of Banking and Finance 1(1):29-54.
3.Andrew L. Johnson and Leon F. McGinnis, 2006.07, An outlier detection methodology with consideration for an inefficient frontier. Department of Industrial and Systems Engineering, Georgia Institute of Technology.
4.Banker, RD, A. Charnes & W. Cooper, 1984, Some models for estimating technical and scale inefficiencies in Data Envelopment Analysis. Management Science 30:1078-1092.
5.Beaver, W. H., 1966, Financial ratios as predictors of failure. Journal of Accounting Research 4(3):71-111.
6.Blum, M., 1974, Failing company discriminate analysis. Journal of Accounting Research 12(1):1-25.
7.Emel, A. B., Oral, M., Reisman, A., and Reha, Y., 2003, A credit scoring approach for the commercial banking sector. Socio-Economic Planning Sciences, Vol. 37, pp.103-123.
8.Farrell, M. J., 1957, The measurement of productive efficiency. Journal of the Royal Statistical Society, Series A, pp. 253-281.
9.Feroz, E. H., Kim, S., Raab, R. L., 2003, Financial statement analysis: A Data Envelopment Analysis approach. Journal of Operational Research Society, Vol. 54, pp.48-58.
10.Golany, B. and Roll, Y., 1989, An application procedure for DEA. Omega 17(3), 237-50.
11.Jim Collins, 2011.01, 為什麼A+巨人也會倒下，初版，台北：遠流出版事業。
12.Joe Zhu, 2009, Quantitative models for performance evaluation and benchmarking using DEA. Springer Science, Department of Management, Worcester Polytechnic Institute.
13.Liu, F.-h. F. and T.-n. D. Hsu., 2004, Least-efficient frontiers of Data Envelopment Analysis-CCR Model. Working Paper.
14.Maria Psillaki et. al., 2008, Evaluation of credit risk based on firm performance. University of Nice-Sophia Antipolis, GREDEG-CNRS.
15.Mensah, Y. M., 1984, An examination of the stationary of multivariate bankruptcy prediction models: A methodological study. Journal of Accounting Research 22(1):380-395.
16.Ohlson, J. A., 1980, Financial ratio and the probabilistic prediction of bankruptcy. Journal of Accounting Research 18:109-131.
17.Paradi, J. C., M. Asmild and P. C. Simak., 2004, Using DEA and worst practice DEA in credit risk evaluation. Journal of Productivity Analysis 21(2): 153-165.
18.Sinkey, J. F., 1975, A multivariate statistical analysis of the characteristics of problem banks. Journal of Finance 30:301-319.
19.Sueyoshi, T., 2003, Mixed integer programming approach of extended DEA-Discriminant analysis. European Journal of Operational Research 152(1):45-55.
20.Troutt, M.D., Rai, A., and Aimao Z., 1996, The potential use of DEA credit applicance systems. Computers and Operations Research Vol. 23, No. 4, pp.405-408.
21.Zmijewski, M. E., 1984, Methodological issues related to the estimation of financial distress prediction models. Supplement to Journal of Accounting Research 22:59-68.