This study presents a new methodology for testing changes in skewness
between time periods (or samples) using the bootstrap method. A Monte
Carlo simulation experiment was conducted to compare the effectiveness
of the bootstrap method with the method suggested by Lau, Wingender a
nd Lau (1989) to test skewness persistence. The results show the boots
trap method to be more powerful than the other method. The bootstrap m
ethod was also used to determine the persistence of skewness in stock
returns. The results show that, in a large percentage of stocks, skewn
ess persists over time.