Global Journal of Human Social Science, E: Economics, Volume 23 Issue 2

© 2023 Global Journals Volume XXIII Issue II Version I 56 Global Journal of Human Social Science - Year 2023 ( )E Impact of COVID-19 on Stock Market Volatility and Forecast using ARIMA and EGARCH -400 -300 -200 -100 0 100 200 300 400 IV I II III IV I II III 2018 2019 2020 Differenced GSECI -400 -300 -200 -100 0 100 200 300 400 IV I II III IV I II III IV I II III IV I 2017 2018 2019 2020 Differenced GSECI Figure 2: Plot of transformed GSECI After first differencing, the series shown in Figure 2, is now mean-reverting. That is, the GSECI is now integrated of order one, I (1). b) Estimatesof ARIMA Model ARIMA informs that the series in question has gone through an integration process before beingused for any analysis. Before deciding on the appropriate ARIMA model to be used for the data sequence, Figure 1 presents the correlogram plots of the differenced GSECI which indicates thelevel of significance of the Q- statistics of a specific set of lags from one to inform our decision on the ideal ARIMA model. exponential decay till lag 4 where the ACF extends beyond the confidence interval bounds and continue decaying exponentially. There exists a slight similarity between the ACF and the PACF (see Appendix 2 for the correlogram). Since the pattern of the ACF and PACF looks the same, we can conclude having a set of The decision on the appropriate lags for the ARIMA model is to arrive using the Autocorrelation Function (ACF) and the Partial Autocorrelation Function (PACF). The autocorrelation of the first difference of the Ghana stock exchange composite index shows that at the lag one, the ACF is significant and shows an