CMStatistics 2018: Start Registration
View Submission - CFE
Title: A Bayesian commodity style-integration framework Authors:  Ana-Maria Fuertes - Cass Business School - City University London (United Kingdom)
Nan Zhao - Cass Business School, City, University of London (United Kingdom) [presenting]
Abstract: The literature abounds with multiple long-short investment styles designed to capture the commodity futures risk premia of which momentum, hedging pressure and term structure are the most well-known. Combining investment styles is strongly motivated a priori because by relying on a composite signal that exploits $K$ lowly-correlated signals, the integrated-style portfolio is likely to capture a larger premium consistently over time. A key decision that an investor pursuing a style-integrated portfolio faces is how to dynamically choose the weights to allocate to each style. A Bayesian style-integration (BI) approach is developed which allows the investor explicitly to account for parameter and model uncertainty. Focusing on the allocation problem of a commodity futures investor that seeks exposure to the well-documented hedging pressure, term structure and momentum styles, the Bayesian integration is confronted with the widely-used nave equal-weight integration (EWI) and the optimized integration approach that obtains the style weights at each portfolio formation time by utility maximization. We find that the proposed BI approach yields a superior Sharpe ratio and certainty equivalent return than the EWI and OI approaches. The findings are robust to transaction costs, variants of the sophisticated OI integration, longer ranking windows, and different economic period analysis.