Identification of calendar anomalies in selected stock exchanges of CEE countries ; Kalendorinių anomalijų identifikavimas pasirinktose Rytų ir Centrinės Europos šalių akcijų biržose

The purpose of this paper is to research the presence of calendar anomalies (Halloween, January, Weekend effects and Santa Claus rally) in selected CEE countries through the period 2010 – 2016. Paper has four main objectives. The first one is to conduct a situation analysis on selected CEE countries market data and its trends, to overview Efficient Market Hypothesis (EMH). The second objective is to review the literature and models on calendar patterns - anomalies and prepare methodology. The third is to conduct empirical research on calendar anomalies in selected CEE countries stock markets. The last objective is to analyze the results on calendar anomalies, identify in which market which anomaly is the strongest, if any. Also, give recommendations for investors depending on the findings. Research methods and data used in this paper: Information required to test anomalies gathering via Bloomberg terminal, analysis of scientific literature. ADF test to check if the data is stationary, regression in OLS and GARCH models. PACF test to determine the presence of autocorrelation at different time lags. The results revealed that Halloween effect is present in VILSE and TALSE indices. January effect is present in Baltic states and SOFIX indices. Santa Claus rally exists in five indices: PX, RIGSE, SOFIX, TALSE, VILSE. Weekend effect results reveal that the effect is present in RIGSE, TALSE and VILSE indices. None anomalies, which were tested was found present in BET, BUX and WIG indices. All anomalies were present in TALSE and VILSE indices.