Jana, R. K., Ghosh, I., Jawadi, F., Uddin, G. S., & Sousa, R. M. (2025). COVID-19 news and the US equity market interactions: An inspection through econometric and machine learning lens. Annals of Operations Research, 345(2), 575–596. https://doi.org/10.1007/s10479-022-04744-x

This article investigates the interactions between COVID-19-related news and the U.S. equity market during the first pandemic wave (January–March 2020), using econometric and machine learning techniques. It examines how global and local COVID-19 fears, measured through daily infection data, influenced 20 U.S. sectoral stock indices. The study divides the sample into two periods: TH-I (January), when infections were mostly global, and TH-II (February–March), when local infections surged. Using Johansen co-integration, DCCA, and nonlinear Granger causality, alongside Gradient Boosting and Random Forest models, the authors find that COVID-19 fears affected sectors differently across time. In TH-I, global fears had limited and mixed effects, while in TH-II, both global and local fears negatively influenced all sectors—particularly automotive, retail, and technology. Predictive accuracy improved in TH-II, reflecting stronger market sensitivity. Overall, the study concludes that local fears became dominant drivers of market volatility as the pandemic escalated.