The Extended Holiday Effect on US Capital Market

Abstract

Studies on the financial markets proved that not all calendar anomalies are persistent in time. Some of them experienced various types of changes, including passing from the classical form to an extended one, with an enlarged specific time interval. This paper approaches the Holiday Effect extended form on the United States capital market. In its classical form, the Holiday Effect refers to abnormal stock returns on a trading day before a public holiday and a trading day after. We study the behavior of stocks returns for a time interval that starts four trading days before a public holiday and it ends four trading days after. In this investigation we employ the daily closing values of four important indexes from the United States capital market: Dow Jones Industrial Average, Standard & Poor's 500, Russell 2000 and NASDAQ Composite. In order to capture the changes experienced in time by the Extended Holiday Effect we analyze the returns of these indexes for three periods: January 1990 - December 1999, January 2000 – December 2009 and January 2010 – April 2020. The investigation revealed, for some trading days from the enlarged specific time interval, returns that were, in average, significant larger or smaller than those of the days outside of this interval. We found especially high abnormal returns on four or three trading days before public holidays and low abnormal returns on one or two trading days after public holidays. The results also suggest that the Extended Holiday Effect was more visible in relative quiet periods than in the turbulent ones and it influences especially the stock returns of small cap companies.