標題: 公司治理、學習效果與股票報酬
Corporate Governance, Learning and Stock Returns
作者: 俞佳德
黃宜侯
Yu,Chia-Te
財務金融研究所
關鍵字: 公司治理;學習效果;股東權力;資訊;效率市場;反收購;Corporate governance;E Index;Learning;Shareholder right;Information;Market Efficiency;Anti-takeover
公開日期: 2016
摘要: 本研究使用Bebchuk et al. (2009) 的邏輯自製E Index,並以Fama-French五因子模型的因子作為控制變數,衡量是否仍有1990到2000年間E Index對超額報酬有解釋力,而其後解釋力消失的「學習效果」的情形。藉由使用Fama-French 五因子模型 (2015)取代原有的Fama and French (1993) 的資產定價三因子模型做檢測的方式,去說明1990到2000年間的異常報酬並非由於遺漏重要風險因子的關係而造成。本研究進一步加入資訊變數來檢測「學習效果」是否會因為投資人能更加容易的取得更豐富的公開資訊或是市場上有更多投資人能理性使用資訊而效果更佳。另外,本研究延伸E Index樣本年間至2014年,且發現 E Index作為一個公司治理指標,在2006年後應被解釋為公司獨立性及公司戰略穩定性的指標而非股東權力大小的指標。
The learning effect is: from 1990 to 2000, we can use the E Index to explan the excess return, however, the correlation disappeares subsequently. We use the method of Bebchuk et al. (2009) to construct the E Index, and use the factors in Fama-French five-factor model as control variables to test whether the learning effect still exists. We use MISPP (the absolute value of MISP, which is the Mispricing Index) as the proxy of learning effect and add the information variables to better describe the learning process. In addition, we construct the E Index from 2007 to 2014 as new sample period and find out that there is a new interpretation in this sample period. The E Index should be a proxy of the independency and the stability of the strategy for the specific company during the new sample period.
URI: http://etd.lib.nctu.edu.tw/cdrfb3/record/nctu/#GT070353917
http://hdl.handle.net/11536/138659
顯示於類別:畢業論文