Volatility Forecasting Performance of GARCH Models : A Study on Nordic Indices During COVID-19

Abstract: Volatility forecasting is an important tool in financial economics such as risk management, asset allocation and option pricing since an understanding of future volatility can help professional and private investors minimize their losses. The purpose of this paper is to investigate the volatility forecasting performance of symmetric and asymmetric GARCH models on Nordic indices during COVID-19. The models examined in this paper are GARCH, EGARCH, GJR and APARCH and the forecasting performance is evaluated by several statistical performance measures. The results of this paper are that the symmetric GARCH(1,1) on average has the worst forecasting performance during a crisis. However, the difference between the predictability of the models is in practice small. The superior forecasting models are the GJR(1,1) and EGARCH(1,1) when forecasting a crisis on Nordic indices according to the evaluation measures. This is because the asymmetric extensions of these models enable them to capture the more present leverage effect in the even tof a crisis.