Time-Series Analysis and Forecasting of S&P500 Index Based on ARIMA and ETS Model

Abstract

The study embarks on an insightful journey into the world of stock indices through the deeper understanding of time series analysis application, especially during the special period of oil price volatility. The goal of this paper is to utilize sophisticated statistical models, including Autoregressive Integrated Moving Average (ARIMA) and Exponential Smoothing (ETS), to extract meaningful information from historical stock index data of S&P500 by Yahoo Finance, enhancing predictive accuracy and operating forecasts to inform strategic decision-making. In addition to the prediction results, the study also compares the forecasts of the two models through some values, and concludes that the prediction effect of the ARIMA model is better than that of the ETS model. The reason behind this result has a lot to do with the processing of the data itself and the fit of the models. For S&P500 during the period of the study, the ARIMA model’s prediction result is better. The insights derived from this analysis are expected to empower investors, researchers, and market analysts with a deeper understanding of the stock index’s past behavior and its implications for future performance.