In recent years, a large volume of empirical evidence has been presented showing that security returns are predictable based on public available information.
All past research typically studied the issue of stock market overreation from the viewpoint of investigating different firms’ performance for a given interval. As a result, these studies are subject to and focus on cross-sectional differences, such as the bid-ask spread, January effect, and other firm-specific factors that may explain the overreaction. To avoid the confounding effects resulting from the cross-sectional differences in individual stocks, this paper used portfolios to investigate the overreaction of markets rather than that of individual stocks. Since the study focused on portfolios to avoid the potential bias due to firm size and the bid-ask spread, the finding of over- or underreaction in portfolios suggests that this stock market is inefficient independent of cross-sectional differences in individual stocks.
To obtain this objective, the best and worst performing days in the portfolio’s history were used to form the winner and loser portfolios. Then, the performance on the subsequent days relative to the initial returns was computed to identify any over- or underreactions in the portfolio.
This paper finds that the Taiwan stock market appears to underreact to good news and overreact to bad news. Though the January and February effects, and the difference in risk can partially explain the profits of losers and winners, the over- or underreaction effects cannot be completely subsumed under them. Market’s underreaction to good news tends to better characterize the return continuation and market’s overreaction to bad news is a proper explanation for the reversal pattern.