ASSESSMENT OF DAY-OF-THE-WEEK EFFECT AND OTHER STOCK MARKET ANOMALIES: ROMANIAN EVIDENCE

Abstract

I examine the existence of calendar anomalies, such as the day-of-the-week effect and January effect, on the Romanian stock market. The day-of-the-week effect was analyzed for the sample period 2002-2022 and three subperiods using a dummy regression. The estimation method for the regression was OLS after testing for normality. The January effect was analyzed for a similar sample period and three subperiods using Wilcoxon rank-sum test. Data were downloaded from Refinitiv Workspace. Even though the results indicate the presence of day-of-the-week and January effects for the Romanian BET index, the conditions on the market modified during the subsample periods. Thursday was the day with the second-lowest risk measured with the standard deviation and the highest return compared to the rest of the week at a 5% significance level. However, Tuesday had a higher return for the subsample period 2011-2022 at a 1% significance level. The January effect was confirmed only for the subsample period 2003-2007, which corresponded to the pre-crisis period and the impact diminished over time, which is a similar result as the ones from the specialized literature. The changing market conditions reduce the possibility of speculating from market anomalies.