Calendar Anomalies and Stock Returns Volatility at Nairobi Securities Exchange: An OLS and GARCH (1,1) Approach

Abstract

Aim: The Kenyan financial market is weakly inefficient due to calendar anomalies explained by stock returns volatility. However, there is a dearth of literature on how the results may vary when different models are applied especially in Africa. Therefore, this study attempted to establish the calendar anomalies in the stock returns volatility using OLS and GARCH (1,1) models. Methods: A descriptive research design was adopted. The daily closing prices data for NSE 20 share index between January 1994 and December 2014 with a total of 5203 observations was used. Results: The results suggest that, the day of the week effect is significant in both models where Friday had the highest returns while Monday had the lowest returns but, the January effect is only explained in the OLS model which disappears in the GARCH (1,1). Conclusion: The coefficients of the volatility equation for GARCH (1,1) are positive, significant and their summation is close to one indicating that the volatility is persistent at the NSE. The study however applied OLS and GARCH (1,1) which have limitations. Recommendation: The study recommend the use of other GARCH models to determine if the findings are robust to different models. JEL Classification: G11; G12; C31; C22