Cross-distributional robustness of conditional weekday effects: evidence from European equity-index returns

Abstract

The paper re-examines the issue of the robustness of the weekday effect. Specifically, by utilizing a quantile regression approach, the homogeneity of observed day-of-the-week anomalies is monitored and tested over different parts of the conditional return distribution. The day-of-the-week effects are measured for conditional returns as well as for conditional volatilities. The model applied accounts for asymmetry in first-order autocorrelation in both moments. The weekday patterns in the returns on the European market index and 18 European country indexes are analyzed for the time period from January 2000 through December 2006. Generally, the sign of the estimated weekday effects in both the conditional mean and volatility seems to be very robust over the return distribution. However, about one half of the country-specific indexes exhibit significant variation or asymmetry in the day-of-the-week coefficients across the quantiles of the conditional return distribution. Only in a few cases, the significant day-of-the-week effect is clearly driven by extreme events.