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

Consistent with Connolly's(1989), (1991)evidence, this study finds tha t sample size and/or error term adjustments render U.S. day-of-the-wee k effects statistically insignificant. In contrast, day-of-the-week ef fects in seven European countries and in Canada and Hong Kong are robu st to individual sample size or error term adjustments, and day-of-the -week effects in five European countries survive the simultaneous impo sition of both types of adjustments. In most countries where day-of-th e-week effects are robust, however, the effects are statistically sign ificant in not more than two weeks out of the month. These findings ar e inconsistent with explanations of the day-of-the-week effect based o n 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 complicat es the international day-of-the-week effect puzzle.