Testing excess returns on event days: Log returns vs. dollar returns |
| |
| Affiliation: | 1. Shanghai Advanced Institute of Finance, Shanghai Jiaotong University, Shanghai 200030, China;2. Department of Money and Banking, National Chengchi University, Taipei 11605, Taiwan;3. Department of Applied Finance, Yuanpei University, HsinChu 30015, Taiwan;1. Centre for Strategic Economic Studies, Victoria University, Melbourne, Australia;2. Asia Pacific Energy Research Centre, Tokyo, Japan;3. School of Economics and Finance, Victoria University, Melbourne, Australia |
| |
| Abstract: | The results of academic and practitioners’ event studies are often translated from excess log returns into excess dollar returns. The prior literature argues for a difference between the statistical significance of excess log returns and that of excess dollar returns. In contrast, we show analytically and using simulations that specifying event study hypotheses in terms of excess dollar returns is equivalent to specifying them in terms of excess log returns. The prior literature’s result was due to a bias in the estimator of expected excess dollar returns, an incorrect assumption that it is approximately normally distributed, and a misapplication of the delta method. |
| |
| Keywords: | Event study Dollar return Statistical significance |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
