Commodities agrícolas do agronegócio brasileiro : análise multifractal e análise da complexidade diante da crise financeira mundial subprime 2008/2009

Abstract

The growth of the world economy, driven by emerging countries, especially China, has generated signi cant changes in the commodities market since 2002. The commodity prices have shown a signi cant increase, reecting the erce conditions of supply and demand for these products, driven by the climatic phenomena that have negatively afected the supply, and by the demand growth rate. The global nancial crisis began in the US market, and eventually turned out the worst global nancial crisis since 1929 (the break of the New York Stock Exchange). The bankruptcy of Lehman Brothers investment bank on September 15, 2008 marks the transformation of the international nancial crisis, after which in Brazil there was a great reduction of international credit, accompanied by a sharp increase of the dollar exchange rate. Considering that the agricultural sector is of fundamental importance to the economic health, being a major investor in environmental and rural technologies, Brazil can not succumb to the idea of a slowdown in this sector, as in 2008 the Brazilian agribusiness represented 36.7% of exports, generating 37% of jobs, and 28% of gross domestic product (GDP). This work investigates the returns asynchrony and the behavior of the cross-correlations for six agricultural Brazilian agribusiness commodities, for the period prior to the global nancial crisis (2006-2009), and after the crisis (2010-2014). The Cross-Sample Entropy method was used for quantifying the asynchrony among the commodity returns series. In addition, the methods Multifractal Detrended Cross-Correlation Analysis (MF-DCCA), Multifractal Detrended Fluctuation Analysis (MF-DFA) and Detrended Cross-Correlation Analysis (DCCA) were used to investigate cross-correlations and auto correlations in the returns series. The results of multifractal analysis show that for all time series, the multifractality decreased after the global nancial crisis, indicating smaller range of the scale invariant fluctuations, except for Cotton, which exhibits precisely the opposite behavior. Based on the obtained results, it can be concluded that the multifractal analysis and the complexity analysis can be useful in the studies of the dynamics of the Brazilian agribusiness, given its importance within the global economic scenario, for adoption of monetary and scal policies by the responsible economic agents, or by the federal government.