The Reexamination of the Overreation Effect in the Taiwanese Stock Market

• Main Authors: Lin Chien Yu, 林倩郁
• Other Authors: Liao Tung Liang
• Format: Others
• Language: zh-TW
• Published: 1999
• Online Access: http://ndltd.ncl.edu.tw/handle/84992429087567453825

碩士 === 逢甲大學 === 企業管理學系 === 87 === The so called "market overreaction" describes a situation where, after certain events result in drastic changes in prices and these changes overshoot that which is predetermined by theory, a rebound phenomena is created to correct the overshoot. Because investors tend to over-emphasize the immediate market condition, investors usually appear to be overly optimistic, or overly pessimistic. Therefore, stock prices usually do not reflect the real reasonable price. Over time, the winner portfolio and loser portfolio will readjust to reflect their true market worth. This paper will focus on the publicly traded companies on the Taiwanese Stock Exchange from January 1981 to August 1998. 25 testing strategies have been designed to test for market overreaction. We uses Holding Period Return (HPR) to calculate the formation period abnormal return, then taking the highest 5% as the winner portfolio and the lowest 5% as the loser portfolio. To test for the presence of market overreaction during the testing period, we used the Stochastic Dominance model and the Extended Mean-Gini Index instead of the t-test. Major findings can be summarized as follows： 1. Employing the Extended Mean-Gini Index, for an investor with a risk aversion parameter of 2, strategies (6,12)、(9,1)、(9,6) and (9,12) showed market over-reaction. When risk aversion parameter is100, strategies (1,12)、(6,12)、(9,1)、(9,6) and (9,12) demonstrated market overreaction. 2. Employing the Stochastic Dominance model, strategies (9,1)、(9,9) and (6,12) demonstrated market overreaction under the SSD (Second-Degree Stochastic Dominance) criteria, whereas strategies (1,6)、(1,9)、(1,12)、(3,3)、(3,6)、(3,12)、(9,12)、(12,6) 、(12,9) and (12,12) showed market overreaction under the SSDR (Second-Degree Stochastic Dominance with Riskless Asset) criteria. 3. Our research results are inconsistent with the findings of King and Hung (1996). This discrepancy can be attributed to : 1) This paper adopted the HPR model, which differs from that used by them. 2) This research employs the EMG and SD Rules instead of the t-test used by them.