Stock market forecasting accuracy of asymmetric GARCH models during the COVID-19 pandemic

Abstract

We propose a new clustering approach for comparing financial time series and employ it to study how the COVID-19 pandemic affected the U.S. stock market. Essentially, we compute the forecast accuracy of asymmetric GARCH models applied to S&P500 industries and use the model forecast errors for different horizons and cut-off points to calculate a distance matrix for the stock indices. Hierarchical clustering algorithms are used to assign the set of industries into clusters. We found homogeneous clusters of industries in terms of the impact of COVID-19 on US stock market volatility. The industries most affected by the pandemic and with less accurate stock market prediction (Hotels, Airline, Apparel, Accessories & Luxury Goods, and Automobile) are separated in Euclidean distance from those industries that were less impacted by COVID-19 and which had more accurate forecasting (Pharmaceuticals, Internet & Direct Marketing Retail, Data Processing, and Movies & Entertainment).