We show how investors' liquidity patterns can provide a common framework to explain autocorrelation of returns and volumes, and some calendar anomalies. The method helps us find new anomalies, and contribute to the explanation of older ones. We uncover a \textquotedblleft festivities effect\textquotedblright\ that is composed of a pre-festivity period of negative returns and relatively low trading activity, and a post-festivity period of positive returns and increased trading activity. We demonstrate this effect for ten countries in the Middle- and Far-East where the main festivities occur every year at a different time of the Western Gregorian calendar. In particular, we consider the Muslim Ramadan and Chinese New Year festivities in these countries.
|Publisher||Department of Applied Mathematics, University of Twente|