[[alternative]]The Comparison of the Predictive Model of Financial Distress on Listed Company in Taiwan

碩士商業教育學系[[abstract]]摘要 當企業發生危機或經營失敗時，不僅企業本身受影響，對企業股東、債權人、員工、往來廠商及銀行等都會帶來極大的衝擊，甚至企業經營失敗對國家整體的資源使用造成相當大的損失。為防止損失產生或避免損失擴大，須能提早察覺企業的危機，則須建立一套財務預警模型來預測企業危機的發生，以協助企業改善現況，不再惡化，並提供訊息讓投資大眾、債權人、管理當局、銀行等調整其決策，以減少損失及資源之誤用。 本研究選取22家交易方式變更為全額交割股及下市之公司為危機公司，再以配對方式選取43家正常公司，選擇危機發生前五年的財務比率為分析之基礎，探討Logit模式是否優於區別分析之區別能力，本研究中實證顯示所選取的21項財務比率，僅總資產週轉率符合常態分配，而Logit模式分析並不受常態性分配的影響，但區別分析卻須符合常態性的檢定，因而本文研究中Logit模式大體上優於區別分析模式的區別能力。[[abstract]]Abstract Whenever a crisis or financial distress occurs to some well-known enterprise, the effect can be a series of misfortunes. Not only can it affect the institution, but also make a dreadful impact on its related shareholders, creditors, stuffs, relation vendors and banks. Accordingly, those enterprise crises usually result in tremendous loss of the country resources. For that reason, this thesis aims to create a predictive model of financial misery, which can forecast the incidence of monetary difficulties. With anticipation, the model could serve as a great help for the management of the company so as to improve its current situation and reduce the possibility of enterprise disaster. Totally, 22 companies have been selected to be examples of try-out whose trading were changed or listing were reclassified, and then sampled with 43 successfully run companies. Both of the predictive models, Logit Model and Discrimiant Analysis, were studied to compare their discriminating ability of crisis occurrences. Furthermore, they were both based on 21 financial ratios of these failed and non-failed firms’ 5 years prior to the financial distress. What’s more important, this report showed that these selected 21 financial ratios were hardly under normalization except the factor of return on asset. As a result, the discriminating ability of Logit model has been found normally superior to that of Discriminant Analysis by this research