本文利用Logistic迴歸建構財務預警模型(financial early-warning models)，試圖了解那些因素有助於更準確地預測台灣傳統產業發生財務危機的可能性，儘可能在公司尚未發生財務危機前，預測公司可能發生財務危機之機率，以收事前預警及風險管理之效。實證分析上，本文利用財務資訊與非財務資訊變數作為解釋變數，試圖篩選出那些財務資訊變數與非財務資訊變數可作為台灣傳統產業財務預警之用。 本文蒐集2000-2008年傳統產業上市上櫃公司之28項財務比率變數（財務資訊變數）與4 項公司治理變數（非財務資訊變數）資料，利用Pearson相關檢定和逐步迴歸消除共線性並選取顯著變數，再設立含財務比率變數的模型一及含財務比率與公司治理變數的模型二，利用Logistic迴歸估計之，以檢視加入公司治理變數後對財務危機預測能力的影響。 實證結果發現，在財務危機發生前一至前三年中，預測財務危機之主要財務資訊變數包括財務結構、償債能力和經營能力指標，而加入公司治理變數後預測準確性提高，此項結果值得投資人作為投資決策之參考。適度降低負債比率及避免更換會計師將改善傳統產業財務結構、增強外部監控與公司治理能力，而提高應收帳款週轉率與現金流量允當比率將提升傳統產業之經營能力與公司價值。
This paper primarily used statistical methods to establish financial early-warning models that made it be possible to predict in advance the probability of some traditional companies experiencing financial distress. This empirical analysis is the first study that attempts to use the financial and non- financial ratios, and this study used Logistic regression models to make a comparison with the financial ratios and the initiation of corporate governance variables in 2000~2008 years. Findings, upon examining the predictor variables for corporate financial distress at one, two, and three years prior to distress, we found that the achievement result of the conventional company had financial structure, solvency, and operating performance are the principal ratio variables. To increase the degrees of predictive accuracy, we employed the corporate governance and investment decisions variables, including the non-financial ratios, the debt outside appropriately and cash flow reinvestment ratio, switched of CPAs, outside monitoring and corporate governance power, receivables turnover and cash flow adequacy ratio. Therefore, this paper may be useful for researchers and practitioners who are focusing on investment decisions, corporate governance, financial risk management and financial forecasting implementation.
