Calendar Effects in Fifty-five Stock Market Indices

This paper examines the calendar anomalies/effects in 55 Stock market exchange indices of 51 countries around the world. The calendar effects which are examined are the turn-of-the-Month effect, the day-of-the-Week effect, the Month-of the-Year effect and the semi-Month effect. The methodology which is followed is the test hypothesis of two unequal data samples with bootstrapping simulated t-statistics. Simultaneously, with the same procedure a seasonality test is applied in order to investigate if more frequent seasonality on expected returns or in volatility is presented. The conclusion is that we reject all calendar effects in a global level, except from the turn-of-the-Month effect, which is present in 36 stock indices and that there is higher seasonality in volatility rather on expected returns, concerning the day of the week and the month of the year effects. The main purpose of the paper is to present a methodology appropriate for data mining which rejects the existence and persistence of main calendar anomalies as the Monday and January effects, while previous methodologies accept them. So this paper presents an alternative approach in the estimation of calendar anomalies and data mining, as well gives some guide notes for financial strategy