透過您的圖書館登入
IP:3.17.74.227
  • 學位論文

企業財務預警模型之實證研究

An Empirical Study on Establishing Precautionary Finance Model of Company’s Performance

指導教授 : 何瓊芳

摘要


本研究試圖建立一套實務上兼顧財務比率、經營管理、非財務資料、關係人交易等徵信資訊,可供企業授信評量使用的預測模型,以作為授信審核的主要考核依據,期許對未來授信判斷上有所助益。為有效控制授信風險,建立一套完善的放款審核機制,才是預防信用風險的根本解決之道,有效之財務預警模型,可幫助企業事先發現自己之危機所在,亦能對金融機構的授信決策,與投資大眾的投資判斷上有所助益。 本研究之樣本資料為國內上市上櫃之危機公司,將危機公司與正常公司以1:2比率配對,樣本區間為2004年至2007年第一季期間,研究樣本的資料來源為台灣經濟新報(Taiwan Economic Journal, TEJ)及公開資訊觀測站(Market Observation Post System),採取產業及規模相仿之正常公司外,本研究以同一產業別,且其危機發生前一年度營收相當之正常公司為基準。 危機公司與正常公司其前三年的財務資料與非財務資料,因樣本變數幾乎皆為非常態分配,故透過無母數統計法平均數差異檢定的Mann-Whitney U檢定,找出危機公司與正常公司有顯著差異之研究變數。 研究結果顯示負債比率、負債淨值比、利息保障倍數、利息保障倍數、流動比率、速動比率、借款依存度、應收帳款週轉率、存貨及應收帳款/淨值、資產報酬率、每股盈餘、稅後淨利率、營業利益率、已實現銷貨毛利率、股東權益率、常續性利益率、營收成長率、營業毛利成長率、營業利益成長率、總資產報酬成長率、現金流量允當比率、現金流量比率、現金再投資比率、支付現金股利、支付董監事酬勞及員工紅利、TCRI 信用評等、應付帳款票據-關係人等二十七個變數,在財務危機發生前一至三年危機公司與正常公司之間皆有顯著差異性。關係人交易方面,財務危機發生前一年進貨外包加工、應付帳款票據及應付融資等具有顯著差異;財務危機發生前二年應付帳款票據、保證之或有負債及工程履約背書保證等具有顯著差異;財務危機發生前三年應付帳款票據、保證之或有負債等具有顯著差異。 建立財務危機發生前一至三年的Logistic迴歸模型進行實證分析。研究結果顯示財務危機發生前一年之財務預警模型有最高的預測績效,正確區別率達85%,為符合現行需求且準確度佳的授信計量審核模型。危機發生前二年整體準確率為76.72%,其中正常公司判別正確率為89.04%,但有較高的將危機公司誤認為正常公司之型Ⅱ誤差,危機公司之判別正確率為55.81%。危機發生前三年之預警模型整體準確率為82.66%,正常公司判別正確率為91.36%,型二誤差為33.72%,危機公司之判別正確率為66.28%。整體而言,前二年與前三年的預警模型達七成之準確性,具有參考價值,為有效之模型。 本研究結果可提供金融機構作為辦理授信業務時之參考。

並列摘要


This research tries to establish a precautionary finance model of company’s performance. The model covers a company’s financial ratio, management, administration efficiency, and non-financial information and transaction records for enterprises to use the model to get some basis on verifying the performance of companies in the future. Number of samples observed from companies with operational crisis and these from normal companies are obtained at 1:2 ratios. Samples cover a time period from 2004 to the first season of 2007. Generally, for companies with operational crisis, the distribution of finance and non-financial data are not normal distribution, so we apply the method of Mann-Whitney U test to find the difference between the crisis-suffering company and normal company. The result of the study shows that differences for variables such as the company’s rate in debt, net liabilities ratio, the interest earning ratio, liquidity ratio, the borrowed money index, surplus per share, after-tax net interest rate, realized gross sales profit rate, net rate of returns, the long-term interests rate, profit growth rate, reinvestment of rate cash flow, cash dividend as well as company’s credit rank of TCRI, cash flow rate, cash available etc., which took place prior to 1-3 years before the first crisis symptom between the failing company and normal company were great. In addition, two groups show different performances in terms of account payable note, and debt payable. By using the extract independent variable to set up financial crisis model, one gets the first Logistic regression model on company’s performance. The model covers a company’s financial ratio, management, administration efficiency, and non -financial information and transaction records. The result of study shows that the model can supremely predict performances for early warning purposes. For the previous year application of the model prior to the financial crisis occurrence, the correct prediction rate can be up to 85%, it is in accordance with the current need and accuracy for a good credit screening system. This result of study can be offered to many financial institutions as a reliable credit screening device.

參考文獻


賴東聖(2005)。建立企業財務預警模型以降低銀行授信風險之研究。中原大學國際貿易學系碩士論文,未出版。
行政院金融監督管理委員會(2007)。http://www.fscey.gov.tw/。
左峻德、朱正中(1998)。東亞金融危機對我國之影響及因應對策。經濟情勢季評論季刊。
Altman, E.I., Financial Ratios,(1968), Discriminant Analysis and the Prediction of Corporate Bankruptcy using Capital Market Data. Journal of Finance, September, 589-609.
Altman, E.I.,(1980), Commercial Bank Lending:Process, Credit Scoring, and Costs of Errors in Lending. Journal of Financial and Quantitative, 15, 813-832.

被引用紀錄


許秀彤(2013)。營運資金管理對財務績效之影響-以台灣集團之企業為例〔碩士論文,淡江大學〕。華藝線上圖書館。https://doi.org/10.6846/TKU.2013.01169
郭亞媞(2012)。現金流量風險值與電子公司危機預 警-Panel Logit Model之應用〔碩士論文,中原大學〕。華藝線上圖書館。https://doi.org/10.6840/cycu201200794
黃裕盛(2016)。建置企業財務危機預警模型之研究〔碩士論文,逢甲大學〕。華藝線上圖書館。https://doi.org/10.6341/fcu.M0317488
邱立成(2009)。電子業財務危機預警分析之實證研究〔碩士論文,國立臺北大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0023-0707200912303400

延伸閱讀