永續經營及維持競爭力是企業生存之重要目標,因此企業者必須要有明確的營運策略,而企業的經營績效與財務危機會反映在財務報表中。本研究以台灣地區上、下市公司各財務變數及非財務變數資料為研究樣本,樣本選取自西元2003~2016年。本研究應用粗集軟體(RSES)方法來建立財務危機預測模型,以萃取出具有關鍵性之核心屬性,並應用於本研究模型實證驗證,以達到最佳的財務危機預測正確率,進而分析出企業可能發生財務危機之前兆,以供企業管理者、投資人、債權人及金融機構等有效地做出相關決策。經粗集軟體約簡結果顯示,在所收集的24項屬性中,財務比率屬性有21項、非財務屬性有3項(公司年資、公司三年內是否更換會計師及公司是否為四大會計事務所簽證)。最後並將本研究實證分析結果相互比較以探討出較佳預測正確率年度。
The most important managerial economics target is company sustainable management and keep competing. Therefore, business operators must possess explicitly strategy then business performance, financial crisis reflect on financial statements. This research is based on the samples of financial variable and non-financial variable from listed company and unlisted companys, they are from 2003 to 2016. The research applies rough set exploration system (RSES) to construct financial distress prediction models to find out crucial core attributes, and applies to this model to prove the financial distress prediction accuracy, in order to achieve the best accuracy rules and annual of financial prediction crisis. To analysis the omen of financial distress, a company may occur for business managers, investors, creditor and financial institutions to make availably decision. This research collects twenty-four attributes, which contains twenty-one financial rates attributes and three non-financial attributes (company seniority, changing accounting firm within three years, and is it top four accounting firm). Finally, the empirical results of this research are compares with each other to find out the best predict correct rate annual and rules.