On the Hybrid of Arima and Garch Model in Modelling Volatilities in Nigeria Stock Exchange

Abstract

This work examined the implementation of combination of the most effective univariatetimeseries model, ARIMA models with the superior volatility models GARCH, in examiningthedaily stocks returns of 2910 observations. Augmented dickey Fuller and Phillips Perrontest were used to check the stationarity of the series. The series were confirmed stationary after thefirst difference. Comparison of  forecasting accuracy of the hybridization between ARIMAModel and Generalized Autoregressive Conditional Heteroscedastic (GARCH)  processes wasdone using the secondary data of All share Index of Nigeria Stock Exchange series obtainedfrom National Bureau of  Statistics and World Bank Statistics Database dated, fromJanuary2012to October 2023.The empirical results of 2910 daily series monthly  revealed that the ARIMA(1,1,1)-GARCH (1,1) model gives the optimum results in modelling the Nigeria Stock exchangereturns compared  to conditional mean model ARIMA (1,1,1).