Revisiting calendar anomalies: Three decades of multicurrency evidence

•We examine the day–of–the–week (DOW), turn–of–month (TOM) and January effect.•The returns during Monday to Wednesday are higher than Thursday and Friday returns.•The returns during January are higher than the returns during rest of the year.•TOM trading days’ returns are lower than the returns during non-TOM trading days.•Emerging currencies show stronger anomalies than advanced and developed currencies.•The calendar anomalies have gradually disappeared in the recent times.

We examine the day-of-the-week, the January, and the turn-of-month effects in developed, advanced and emerging currencies from 1985 to 2014. The returns on Monday, Tuesday and Wednesday are found to be positive and significantly different from zero. The returns on Thursday and Friday are negative and significantly smaller than the returns during first three days of the week. January returns are higher than the returns during the rest of the year. TOM returns are negative and significantly lower than that of non-TOM returns. The calendar anomalies are found to be stronger for emerging currencies compared to advanced and developed currencies. The subsample analysis shows that the calendar anomalies are stronger during the initial subsample and gradually diminish by the last subsample. We also show that for each calendar anomaly, our implied trading strategy can outperform the buy-and-hold strategy in the initial subsamples not so in the last subsamples. Overall, our results indicate that the calendar anomalies have disappeared in the recent times and the markets have become efficient.