Stock Returns and Long-range Dependence

Abstract

This article studies the long-term memory behaviour of stock returns on the Ghana Stock Exchange. The estimates employed are based on the daily closing prices of seven stocks on the Ghana Stock Exchange. The results of the autoregressive fractionally integrated moving average-fractionally integrated generalized autoregressive conditional heteroskedasticity (ARFIMA-FIGARCH) model suggest that the stock returns are characterized by a predictable component; this demonstrates a complete departure from the efficient market hypothesis, suggesting that relevant market information was only partially reflected in the changes in stock prices. This pattern of time dependence in stock returns may allow for past information to be used to improve the predictability of future returns.