Macroeconomic attention and oil futures volatility prediction

Abstract

This paper mainly checks whether the macroeconomic attention indices have valuable information to predict the oil futures volatility. Results show that macroeconomic attention indices are able to predict the oil futures volatility. In addition, based on several dimensionality reduction methods, we find that the scaled principal component analysis (SPCA) model has better predictive performances than other dimensionality reduction methods. Especially, the least absolute shrinkage and selection operator method (LASSO) has the best predictive performance. During the COVID-19 period, LASSO model with the macroeconomic attention indices can still have superior performances. This paper tries to show new evidence based on macroeconomic attention indices for oil market volatility.