Can commodity returns forecast Canadian sector stock returns?

Abstract

Using a wide range of commodities, we provide some evidence that commodity returns can forecast eight Canadian sector equity returns out-of-sample. In particular, there is some evidence that the recently developed bagging method can improve forecast accuracy relative to the benchmark and performs well compared to forecast combinations. From an economic gains perspective, forecasting sector returns provides certainty equivalent gains in a sector rotation strategy. We also model the impact of transaction costs upon economic value and find that gains can be generated when transaction costs are low.