本研究擬以統計分析針對影響公司治理的變數及股票報酬風險進行探討,期望藉由分析結果,提供投資大眾對於公司治理變數與股票報酬風險程度關聯性的參考依據,期能瞭解在不同股票報酬風險程度中可以選擇合適的投資標的而得到較高的報酬,將可能引發之損失降至最低。 依據國外實務上的經驗,公司治理良好的上市企業,會有較好的股票表現,而投資大眾也多數認同公司治理愈好,經營績效也會愈好,股票報酬風險也會隨之降低。本研究利用K-Means集群分析、Two-Step集群分群及迴歸分析探討公司治理與股票報酬風險程度的關聯性,以K-Means分群法及Two-Step分群法依報酬率變異數區分股票報酬風險程度,然後高低股票報酬風險程度區分結果個別進行迴歸分析,以選擇合適的分群方式做為股票報酬風險程度分類之參考,提供未來研究之參考。
The aim of this research is to measure the level of investment risk to choose the subject and get higher investment returns by using the statically analysis approach. In this thesis, the above factors compose the several corporate governance variables and the risk of stock returns so that the corporate governance and stock returns are considered and used for assessing the investment risk for investors. Based on the proposed approach the investors may reduce the investment risk to an acceptable risk level so as to minimize the possible risks and possible loss in terms of maximizing the profits. According to the experience in practices, the listed companies with better corporate governance are also with better stock performance too. Moreover, the companies with better performance always have lower risk of stock returns for the investors. In this study, we applied K-Means clustering analysis for finding the relationship between stock revenue risk and corporate governance variables. The clustering technique and regression analysis are applied respectively. Clustering technique is to distinguish the listed companies with high risk or low risk to the stock returns and K-Means cluster analysis has been used. While the two clusters of companies are clarified, the two groups are analyzed by using regression approach individually. To understand the relation between stock revenue risk and corporate governance and choose best clustering results as a reference for the level of risk classification could be the reference for future research.