外部監控、代理問題與信用風險模型之實證研究

本文利用Logistic迴歸建構非集團業信用風險模型（Credit Risk Models），考慮代理問題，找出影響公司績效之顯著變數，期能在公司未發生財務風險（Financial Risk）前，預測出可能發生之機率，達事前預警與風險管理（Risk Management）之效。實證分析上，採用K-S檢定、M-U檢定與Logistic迴歸分析。實證結果發現，當董事長兼任總經理，降低代理問題提高預測準確率，可以降低發生財務風險的機率，值得提供投資人決策參考；不論董事長是否兼任總經理，非集團企業宜降低每股現金流量、重視會計師保留意見、避免會計師（CPAs）更換，外部人會計師之監控能力增強，可提高公司治理能力。因此本文之實證價值與管理意涵為非集團業重視公司治理，可降低信用風險發生的可能性。

關鍵字

外部監控；公司治理；信用風險模型；財務風險管理

並列摘要

This paper primarily used statistical methods to establish logistic credit risk models on non-groups companies, discussed the agency problem on whether management and ownership belong to the same person. That made it be possible to predict in advance the probability of non-group companies experiencing financial risk. Empirical analysis, the author built up a credit risk model using K-S test, M-U test and logistic regression model. Finding, In the case where management and ownership are combined, there had no agency problems and increased the degrees of predictive accuracy, empirical results suggested that the non-groups companies reduce the cash flow per share, pay attention to CPA gives a qualified audit opinion, avoid switched of CPAs, and the outsider monitoring and corporate governance capacity have to be enhanced. Implications, the non-groups companies pay attention to corporate governance could reduce credit risk.