Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/18790
標題: 從財務指標觀點建構訴訟預警模型-羅吉斯迴歸之應用
Financial Measures, and the Prediction of Litigation Presence-A Logit Regression Approach
作者: 吳怡瑩
Wu, Yi-Ying
關鍵字: Operation Crisis;經營危機;Litigation Risk;Logit Regression;起訴風險;羅吉斯迴歸
出版社: 會計學研究所
摘要: 
自2001年美國安隆(Enron)破產事件後,接連發生的財務舞弊案件,令投資大眾及債權人等重大的損失,亦使得會計專業界面對愈漸強大之訴訟壓力及法律責任。故為降低起訴企業經營失敗之影響,本研究以傳統財務指標結合羅吉斯迴歸(Logit regression)工具來建構高訴訟風險企業之預測模型,以期協助會計師事務所及投資大眾警覺潛在體制不良之企業,進而作為相關決策之參考依據。
本研究之實證結論顯示,營運現金支付力、資金運用流動性、經營績效等三個構面與企業被起訴風險呈現負相關;而償債能力及應收帳款回收能力等構面則與企業被起訴風險呈正相關。另短期資金適切性與獲利能力構面則是與企業被起訴風險無顯著之相關性。由此結果顯示企業經營投資管理不當、經營績效降低、資本結構持續惡化、營運資金之充裕度不足、喪失清償能力,是影響企業發生經營危機及面臨起訴風險之關鍵要素。另就發現高訴訟風險企業在財務報表上之特徵大致可歸納為:營運現金流量比率呈急速下降、高度運用財務槓桿,不良銷貨品質充斥等現象。
此外,本模型總正確預測率在訓練集或驗證集,皆是以起訴前第1季為最高,分別為86.67%及77.27%,可據以得知在最逼近起訴時點能提供最強烈的警訊。故本研究所建立之企業訴訟預警模型,有其預測效力並可簡便的查覺起訴企業問題徵兆所在。

Since Enron collapsed in 2001, financial scandals increasing in numbers have not only imposed immense losses on investors and creditors, but also led accounting professions to face greater risks in lawsuit. So as to reduce litigation risks arising from operational failures by enterprises, this study aims to establish a prior alarm system for lawsuit through an application of logit model along with financial measures. Hence, it is expected that with system's assistance both accountants and investors could conscious the potential threats stemming from incompetent corporations at their early stage.
The empirical results show that three-category risk factors such as cash flow from operating, capital application and operation performance are negatively related to litigation presence. However, solvency and accounts receivable collection have a positive relation. It is surprisingly found that short-term capital flow and profitability is not coefficients significant. Accordingly, it can notify if enterprises with improper investment management, poor operation performance, inappropriate capital structure, insufficient working capital and insolvency would potentially bring about operational crisis or even lead to lawsuit.
In addition, the highest classificatory accuracy of the training and test sample is found in the first season which is classified correctly 86 percent and 77 percent respectively. A better warning is evident when the time is getting closed to litigation occurrence. Therefore, the litigation prediction model constructed in this study may provide a helpful assistance in identifying the occurrence of lawsuit.
URI: http://hdl.handle.net/11455/18790
Appears in Collections:會計學系所

Show full item record
 
TAIR Related Article

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.