Macroeconomic Indicators and U.S. Financial Market Volatility: A Comparative Time Series Forecasting Study

Abstract

Financial market volatility works as a key indicator of market risk. It plays a significant part in investment decisions. Existing studies pay main attention to volatility modeling or asset pricing, while the predictive contribution of macroeconomic indicators to VIX forecasting are kept still insufficiently explored. This study aims to examine macroeconomic information improving the forecasting accuracy of financial market volatility. It intends to compare the performance of different time-series models based on quarterly data from 2000 Q1 to 2025 Q4. Furthermore, it selects the CBOE Volatility Index (VIX) as the volatility measure, while incorporating GDP, CPI, and UNRATE as macroeconomic indicators. After RMSE, MAE, MAPE, and the Diebold-Mariano test evaluating the forecasting performance, ARIMA, ETS, ARIMAX, VAR models are constructed. The results show that ARIMA achieves the greatest out-of-sample forecasting performance, whereas macroeconomic variables provide limited additional predictive value in this study. With findings obtained, it is suggested that historical volatility information is important to short-term VIX forecasting. This study provides implications for risk monitoring and financial decision-making. However, the analysis still has limitations, including quarterly data frequency and a restriction on the selection of macroeconomic variables. These demerits can be extended in future research.