标题: | 从行为财务观点探讨超额股票报酬及经济附加价值间的关系 The Relationship between Excess Stock Return and EVA from the Behavioral Finance Perspective |
作者: | 陈英茵 Chen, Ying-Yin 王淑芬 Wang, Sue-Fung 财务金融研究所 |
关键字: | 行为财务;因果分析;经济附加价值;Behavioral Finance;Causality test;EVA |
公开日期: | 2010 |
摘要: | 效率市场假说与行为财务学的观点对于市场的资讯反应有所不同,前者认为企业经营资讯会充分反映在股价报酬,而后者则认为股价报酬具有先行性之资讯效果。从过去的文献发现企业的每股盈余(EPS)在效率市场假说之下的资讯含量比较高, 然而在2002 年行为财务学的崛起,发现企业的获利行为与市场的表现并不完全一致。文献上 对于EPS 与经济附加价值(EVA)何者具有较高的资讯含量一直没有定论。有鉴于此,本研究以美国市场的制造业为样本进行实证研究,并以股价报酬与两种超额股价报酬的代理变数分别利用纵横资料回归(Panel 回归)都发现EVA 对这三种股票报酬的资讯内涵较EPS 高,并进一步透过因果关系模型分析,其结果发现EPS、EVA 都与股票报酬间呈现反向因果关系,尤其EVA 相对更为显着,此符合行为财务学的论点。 Efficient market hypothesis (EMH) and behavioral finance have different point of views on the market’s reaction to information. The former believes that business operating information is fully reflected in stock returns, whereas the latter states that stock returns have price leadership over business information. Previous literature shows that earnings per share (EPS) is more informative under EMH; however, after the blooming of behavioral finance in 2002, recent research has shown that corporate operating profitability is not at all consistent with the market perception. Literature has not been conclusive on relative information content for EPS and EVA. This study explores the empirical data of manufacturing firms within the US market. Total stock returns and two excess stock returns are first used as proxies through panel regression analysis, respectively. Results show that EVA is more informative than EPS for all three stock returns. Furthermore, a causality test in performed, revealing that both EVA and EPS present reverse causalities to the three stock returns, and EVA is relatively more significant. These results are consistent with the perspective of behavioral finance. |
URI: | http://140.113.39.130/cdrfb3/record/nctu/#GT079839506 http://hdl.handle.net/11536/48084 |
显示于类别: | Thesis |
文件中的档案:
If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.