2010

Momentum, the tendency of recent winner stocks to continue to rise and recent loser stocks to continue to fall, is one of the most puzzling asset pricing anomalies in modern finance. The recent boom in commodity futures investments has sparked renewed interest from both academia and industry in momentum investment strategies. This thesis proposes and examines the performance of three novel momentum-based active investment strategies in commodity futures. Conventional momentum strategies rely on 12 months of past returns for the formation of investment portfolios. First, this thesis proposes a more granular strategy termed 'microscopic momentum‘, which decomposes conventional momentum into single-month momentum components. The novel decomposition reveals that a microscopic momentum strategy generates persistent economic profits even after controlling for sector-specific or month-of-year commodity seasonality effects. Furthermore, we find that all 12 months of past returns play an important role in determining the conventional momentum profits. Second, for the first time in the literature, we document a consistent reversal pattern in commodity momentum profits. Combining the observed reversal pattern with the momentum signal, the strategy in the second study significantly outperforms conventional strategies. The profitability of the proposed strategy cannot be explained by standard asset pricing risk factors, market volatility, investors‘ sentiment, data- mining or transaction costs, but appears to be related to global funding liquidity. Furthermore, the proposed investment strategy in commodity futures may be employed as a portfolio diversification tool.