Modelling and forecasting S&P 500 stock prices using hybrid Arima-Garch Model

Abstract

Abstract The S&P 500 is a bellwether and leading indicator for the economy as well as the default vehicle for passive investors who want exposure to the U.S. economy via index funds. Since 1957, the S&P 500 has performed amazingly, outpacing other leading asset classes such as bonds or commodities. This study seeks to develop an appropriate ARIMA model that best fit the monthly stock price of S&P 500 for a period of 17 years, 2001–2017, thus make a short-term forecast in a way to give an overview and help the investor or portfolio manager in decision making. EViews software was used to run the analysis of the data. Our analysis involved 2-step procedure, which were identifying ARIMA model then fitting GARCH (1,1) into the model. As a result, ARIMA (2,1,2)-GARCH (1,1) model was found to be the best model for forecasting the S&P500 stock prices. The research findings indicate that a dynamic forecast gave a better result compared to a static forecast.