An analysis of Brazilian agricultural commodities using permutation – information theory quantifiers: The influence of food crisis

Abstract

Abstract We analyze predictability of selected Brazilian agricultural commodity prices (cotton, sugar, coffee, soybean and cattle) during the period 1997–2018 which encompasses the periods of agricultural commodities prices spikes between 2007 and 2011. We use information theory based methods entropy-complexity plane (HC) and entropy –Fisher information plane (HF), where Shannon entropy, statistical complexity and Fisher information measure are calculated for empirical Brandt-Pompe probability distribution of ordinal patterns. We find that coffee market shows lowest predictability (highest efficiency), while pork market shows highest predictability (lowest efficiency). By analyzing temporal evolution of efficiency index derived from Shannon entropy and Fisher information measure we observe that coffee and soybeans price series exhibit high and stable informational efficiency for all analyzed period, the efficiency of sugar market shows steady increase, while the efficiency of cattle and cotton market first decreases (until the crisis) and then increases in the post crisis period.