Risk Premia in the Commodity Market

Abstract

Employing a broad cross-section of commodities, we investigate the risk premia estimates for a large set of tradable and nontradable factors using state-of-the-art machine learning methodologies that account for omitted variable and measurement error biases in asset pricing. We show that momentum factors from equity, bonds, and currencies are priced in the cross-section of commodity returns, designating a link between commodity and other financial markets. Finally, the cross-sectional commodity pricing ability of news-based uncertainty and volatility factors with a negative risk premium suggests that commodities are safer at bad times.

Keywords: Commodities; Factor premia; Observable factors; Nontradable factors;PCA; RP-PCA; SPCA

JEL Classification: G11;G12