A COMPARISON OF GARCH-TYPE MODELS FOR VOLATILITY MODELLING IN THREE DIFFERENT SECTORS OF MARKETS

Abstract

This study examines the performance of various volatility models and distributional assumptions in modelling financial time series data from the Nigerian market. Specifically, the research evaluates the fit of different distributions, Normal, Student’s t, Generalized Error Distribution (GED), and Skew-t, within volatility models, including ARCH, GARCH, and EGARCH, to capture the time-varying volatility of nine selected securities. The performance of these models is assessed using three key performance metrics: Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and log-likelihood. The results indicate that the EGARCH model with the t-distribution provides the best fit for most securities, outperforming the other models in terms of model selection criteria. While the EGARCH model with the Skew- distribution is slightly less effective, it still performs well in comparison to the other models. Overall, the findings highlight the superior ability of EGARCH with the t-distribution to model financial volatility in this context, making it the most robust model for forecasting and risk management in the Nigerian financial market. This study contributes to the growing literature on volatility modelling by providing empirical evidence on the effectiveness of different distributional assumptions in emerging markets.