運用財務比率分析企業財務危機之研究

近年來受經濟全球化的影響，資訊科技及商業貿易蓬勃發展，企業在擴大規模的期間，面臨的挑戰與風險將越趨嚴峻，導致企業的生存更加困難，因此建構嚴謹的信用預警模型，提早防範企業財務危機的爆發，可以保障投資人、股東、企業經理人與銀行等專業機構等之利益。台灣於2013年正式採用IFRS國際會計準則，為了消除其他外在控制因素的影響，考量樣本的取樣須遵循財務報表的編制符合同個原則，故本研究以2015年至2019年間台灣陷入財務困境的上市櫃公司為研究對象，選取非金融產業的42家危機公司，再選入同會計年度、同產業且規模相近的84家正常公司，進行兩者比例為「1：2」的配對。將七大財務構面下之20項財務比率作為解釋變數，利用共線性檢定與逐步迴歸分析法篩選具有預測能力的因子，建立Logistic模型與區別分析模型。實證結果表示，危機發生前一年有四項財務變數與財務危機具顯著相關，危機發生前兩年有五項財務變數與財務危機具顯著相關；在預警模型的部分，Logistic迴歸模型的預測能力皆優於區別分析模型；對於研究樣本取樣時間點而言，危機發生前一年的模型預測力較佳。

關鍵字

財務危機；財務比率；財務預警模型； Logistic模型；區別分析

並列摘要

In recent years, under the influence of economic globalization, information technology and commercial trade are booming. During the period of expanding the scale of enterprises, the challenges and risks faced by enterprises will become more and more severe, which will make the survival of enterprises more difficult. Therefore, build a rigorous credit model and prevent the outbreak of financial crisis of enterprises. Early prevention of financial crisis can protect the interests of investors, shareholders, managers, banks and other professional institutions. In 2013, Taiwan formally adopted IFRS international accounting standards. In order to eliminate the influence of other external control factors, the samples must follow the principles that the preparation of financial statements conforms to the contract. Therefore, this study selects 42 crisis companies in non-financial industry from Taiwan's Listed Companies in financial distress between 2015 and 2019, and then selects them into the same accounting year and the same industry 84 normal companies with similar scale were matched with a ratio of "1:2".Taking 20 items of financial ratios under the seven financial indicators as explanatory variables, using collinearity test and stepwise regression procedure to screen the factors with predictive ability, and establishing logistic model and discriminate analysis model. The empirical analysis results show that four financial variables are significantly related to financial crisis in the year before the crisis, and five financial variables are significantly related to financial crisis in the two years before the crisis; in the part of financial distress prediction model, the prediction ability of logistic regression model is better than that of multiple discriminate analysis model; in the aspect of model comparison, the model prediction ability of the first year before the crisis is better.