Probability weighting in commodity futures markets

Abstract

Probability weighting refers to the behavioral bias in which irrational investors have a propensity to overweigh small probability tail events. In this study, we empirically investigate the asset pricing implications of probability weighting in commodity markets. We find that commodities with a high probabilityweighting value significantly underperform their low-value pairs by 11% per annum. Neither conventional commodity risk factors nor existing characteristics explain this predictability. The predictability is more pronounced when arbitrage constraints are more binding. Commodities with a high probabilityweighting value also attract excess demand from non-commercial traders. Collectively, these findings support a prospect theory-based explanation for the cross-section of commodity returns.