Direction-of-change forecasting in commodity futures markets

This paper examines direction-of-change predictability in commodity futures markets using a variety of binary probabilistic techniques. As well as traditional techniques, we apply Variable Length Markov Chain (VLMC) analysis, an innovative technique popularised computational biology when predicting DNA sequences (Bühlmann & Wyner, 1999). To the best our knowledge, this is first application VLMC finance. Our results show that both and technical analysis methods provide strong returns, with annualised mean returns approximately 8%, substantially higher than passive long strategy. suggest short-term learning effect present commodities market which can be exploited forecasting