Abstract
We analyze the market efficiency of 25 commodity futures across various groups metals, energies, soft commodities, grains and other agricultural commodities. To do so, we utilize the recently proposed Efficiency Index to find out that the most efficient among all of the analyzed commodities is heating oil, closely followed by WTI crude oil, cotton, wheat, and coffee.
On the other end of the ranking scale we find live cattle and feeder cattle. The efficiency is also found to be characteristic for specific groups of commodities, with energy commodities being the most efficient and other agricultural commodities (composed mainly of livestock) the least efficient groups.
We also discuss contributions of long-term memory, fractal dimension and approximate entropy to the total inefficiency. Last but not least, we come across the nonstandard relationship between the fractal dimension and the Hurst exponent.
For the analyzed dataset, the relationship between these two variables is positive, meaning that local persistence (trending) is connected to global anti-persistence. We attribute this behavior to specifics of commodity futures: they may be predictable over a short term and locally, but over a long term they return to their fundamental prices.