Stock Price Volatility Modelling with Regimes in Conditional Mean and Variance
Abdulganiyu Salami
*
Department of Economics, Federal University of Lafia, Nigeria.
Timilehin Olasehinde
Department of Economics, Federal University Oye Ekiti, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
This study examined the presence and nature of volatility in the Nigerian Stock market. Through a graphical presentation of the Nigerian stock prices, it was observed that there exist two regimes of volatility clustering between the periods of 1985M6 to 1999M12 and 2000M1 to 2018M6. Employing a regime covariate autoregressive (AR-X) with an exponential GARCH model, that allows for a shift in intercept, it was found that the second regime, 2000 M1 to 2018 M6, is more volatile, and that modelling of Nigerian stock market requires a technique that considers more than one regime of volatility clustering. Consequently, the study recommends that local and foreign investors take into consideration the high volatility of the recent Nigerian stock prices in making their investment decision and that policymakers take cognizance of the volatility in designing macroeconomic policies.
Keywords: All share index, volatility, clustering, exponential GARCH, stock market.