| 摘要: | 過去文獻進行危機預測時並未對於危機公司進行詳細區分，因此所得出之結果存在一定偏差，故本研究主要目的在於區分危機公司之型態及其適用變數等，變數之選用分為兩種，分別為財務重點專區之七大變數及Altman所設立之Z1~Z5模型，本研究樣本取自台灣經濟新報及證券投資人及期貨交易人保護中心經刪除特殊產業、興櫃、公開發行公司後樣本共185間，研究結果顯示危機公司分為三種型態，分別為先舞弊後危機、先危機後舞弊、純危機公司，其中先危機後舞弊公司因樣本數過少之緣故，因此不具代表性，先舞弊公司及危機公司則證實確實存在差異，先舞弊公司之趨勢於T-1年以前較難發現財務惡化之跡象，但於T-1年時，先舞弊公司之財務狀況明顯下滑，惡化程度會超過危機公司；危機公司則於危機年前呈現持續惡化狀態，並且隨著危機年接近，惡化程度會更加明顯。因此若不將其區分則容易導致危機預測之結果失真，且會隨著預測之期間愈長，失真的可能性愈高。;In the past, the literature did not make a detailed distinction of crisis companies when conducting crisis predictions. Therefore, the results obtained have certain deviations. Therefore, the main purpose of this research is to distinguish the types of crisis companies and their applicable variables. The selection of variables is divided into two types, Financial key Area and the Z1~Z5 model established by Altman. The sample of this research is taken from the Taiwan Economic Journal and Securities Investor and Futures Trader Protection Center.After deleting special industries, counters, and public offering companies, there are a total of 185 samples. The research results show that crisis companies are divided into three types, namely, fraud first and then crisis, first crisis and then fraud, and pure crisis company. Among them, the first crisis and then fraud companies are less representative due to the small number of samples. The type of fraud first company and the pure crisis company confirm that there are indeed differences. The trend of the first fraud company is difficult to find signs of financial deterioration before T-1 year, but in T-1 year, the first fraud company′s financial situation has declined significantly, and the degree of deterioration will exceed crisis company. Crisis companies showed a state of continuous deterioration before the crisis, and as the crisis year approaches, the degree of deterioration will become more pronounced. Therefore, if they are not distinguished, the results of crisis prediction are likely to be distorted, and the longer the prediction period, the higher the possibility of distortion. |
