Research of the Calendar Effects in Stock Returns

In this article we investigate the problem of detection of the statistically significant dependences of stock trading return, which occur in particular days of the month and which could be important for creating profitable investment strategies. This problem is formulated as two hypotheses, stating that the stock trading return of the last five days of the month is greater than the average total monthly return, and the return generated over the first half of the month is significantly larger than that of the second half.. By using the advanced methods of statistical analysis we researched the indications of these calendar effects for 24 stocks of the Vilnius stock exchange. The investigation did not fully confirm any of the hypotheses, but found out strong relation of risk level to the researched periods of the month. We explored the dependency of this effect to the volatility and volume of the traded stocks. The research results revealed that stocks of small and moderate volume have high volatility on the last days of the month, and the stocks of high volume have high volatility on the first part of month.