INTERNATIONAL EVIDENCE ON THE ROBUSTNESS OF THE DAY-OF-THE-WEEK EFFECT

Consistent with Connolly's(1989), (1991)evidence, this study finds that sample size and/or error term adjustments render U.S. day-of-the-week effects statistically insignificant. In contrast, day-of-the-week effects in seven European countries and in Canada and Hong Kong are robust to individual sample size or error term adjustments, and day-of-the-week effects in five European countries survive the simultaneous imposition of both types of adjustments. In most countries where day-of-the-week effects are robust, however, the effects are statistically significant in not more than two weeks out of the month. These findings are inconsistent with explanations of the day-of-the-week effect based on institutional differences or on the arrival of new information. Thus, in the absence of other potential explanations already dismissed by Jaffe and Westerfield (1985), evidence in this study further complicates the international day-of-the-week effect puzzle.