Testing Calendar Effect on Nigerian Stock Market Returns : Methodological Approach

In this study we test for the existence of calendar effect in Nigerian stock market returns. The data utilised comprised of the daily All Shares Price Index returns for a period of 1339 sampled days ranging from 19th April 2005 to 30th September 2010. In testing for calendar effect in the Nigerian bourse (stock exchange) for the period of study, we utilized relevant descriptive statistics such as mean, standard deviation, skewness, kurtosis and Jaque-Bera (JB) test to enable us understand and compare the unique statistical properties of the stock returns utilized. We also subjected the daily stock returns to stationary tests using Augmented Dicker-Fuller (ADF) and the Philip-Per ron (PP) Unit root test, while the data analysis was performed using the multiple ordinary least square regression (OLS) techniques in testing for the day of the week and month of the year effect. Our findings revealed that Monday Thursday and Friday are associated with negative market returns (R) while Tuesday and Wednesday are associated with positive market returns (R) and that equity traders can make abnormal gains by trading on Tuesday in Nigerian stock market since Tuesday was positive and statistically significant. On the other hand, our monthly calendar test results shows that February, March, April, May and December were consistently associated with negative market returns (R) while January, August, September, October and November were associated with positive market returns (R). In the case of June and July there was mixed signs (findings)