Is there a relationship between day of the week, months, holidays and returns on the Icelandic stock market?

Abstract

This study examines the relationship between the day of the week, months and holidays and returns of the Icelandic stock market from January 1993 to May 2003. Empirical tests were performed to test if the day of the week did affect returns, if the month of the year did affect returns and if returns had been abnormal before and after holidays. Two methods were applied in this study, Kruskal-Wallis test and linear regression. The findings were that there is not a statistically significant relationship between the day of the week and returns, and no significant relationship between months of the year and returns. However, there is a statistically significant relationship between returns and days the last trading day before New Year and the first trading day after New Year, and there is a significant statistical relationship between returns and days the first trading day after Easter.