| 摘要Abstract | 本研究以 2003 年至 2005 年作為研究期間，市值前百大企業作 為研究對象，探討加入公司治理指標建構信用評等預測模型之影響。 許多公司治理的變數須人為主觀之判斷，而研究方法中屬類神經網路 最能模擬人類思考模式，在本研究模型整體效度（擊中率 90%），內 部效度（擊中率 89.29%）、外部效度（擊中率 88.57%），類神經都 得到了不錯的結果，表示該模型之預測效果能提供外部關係人未來將 模型應用於樣本外之企業風險預測。實證分析的結果顯示，在最佳演 化代數及基因選取組合下，基因演算法篩選出之九個關鍵變數中，公 司治理指標佔有 5 個。換言之，公司治理變數佔全體關鍵變數組合接 近一半甚至以上之比例，顯示公司治理的確是評估企業風險上不可忽 視之警訊指標。加入公司治理指標建構模型，的確能提升整個模型考 量之完整性，以及對信評結果預測及分類的能力。Our research aims to discuss whether the incorporation of corporate governance variables of non-financial information could enhanced the prediction power to results of credit rating.Our empirical investigation reveals that after corporate governance indicators been incorporated into the model which is constructed by multivariate discriminate analysis to screen and select key variables, hit ratio achieves 89.52 much better than what we get simply by using financial ratio without incorporating corporate governance indictors and that would boost the preciseness and integrity we considered. Furthermore, by using Genetic Algorithm, we extract variables having heavy impact to credit ratting result, and find that more than half of variables belong to corporate governance indicators, and that means corporate governance really important information source of business evaluation. |