考慮總體環境不確定性下之財務危機預測模型-以台灣上市公司為例

Abstract

目前文獻最常考慮影響到公司財務困難的變數大都取自公司之財務比率與會計資訊，而忽略總體環境不確定性對於公司發生財務困難的影響，本研究除了採用台灣上市公司之財務資訊，同時採用不同的市場與總體環境變數探討總體經濟變化在評估公司發生財務困難的機率估計上有何影響。由於Logit 模型已普遍被使用在台灣的財務危機預警模型中，本研究採用Logit model (羅吉斯模型) 和 Discrete-time hazard model (離散危險模型)來評估公司發生財務困難之機率並進行比較。實證結果發現，考慮財務比率、市場與總體經濟變數後之discrete-time hazard model預測公司發生財務困難之能力優於傳統的Logit model。此外，本研究採用Accelerated failure time model (加速時間衰敗模型)，進而探討Exponential、Weibull、Lognormal 和 Log-logistic分配下，財務比率、市場與總體經濟環境變數對於公司存活期間的影響，實證結果發現，Lognormal distribution為存活期間分析之最適分配。

This study argues that hazard models are more appropriate for forecasting bankruptcy than the logit models used previously. This study describes a simple technique for estimating a discrete-time hazard model with a logit model estimation program. Other than the accounting ratios used in previous models, several market-driven variables are considered to estimate bankruptcy probability, including market size, excess stock return and market return of valued-weighted stock index. Since the macroeconomic uncertainty might have some influence on bankruptcy probability, this study also selects several business indicators in our model. We find that half of the accounting ratios that have been used in previous models are not statistically significant bankruptcy predictors and the market return of valued-weighted stock index is strongly related to bankruptcy probability. Moreover, this study proposes a model that uses a combination of accounting ratios, market-driven variables and macroeconomic variables to produce more accurate forecasts than alternative models. To comprehend our bankruptcy analysis, we also construct an Accelerated Failure Time model with Exponential, Weibull, Lognormal and Log-logistic distributions and find that Lognormal distribution is the best-fitting distribution in our duration analysis.