Do the global grain spot markets exhibit multifractal nature?

Xing-Lu Gao,Ying-Hui Shao,Yan-Hong Yang,Wei-Xing Zhou

DOI: https://doi.org/10.1016/j.chaos.2022.112663

2022-01-01

Abstract:The multifractal nature of agriculture-related stocks and derivatives has been extensively studied. However, due to the scarcity and lack of representativeness of available data, the complex behaviors of the global grain spot markets are less researched. We perform multifractal detrending moving average (MF-DMA) analysis on the daily data of the Grains & Oilseeds Index (GOI) and its five sub-indices of wheat, maize, soyabeans, rice and barley. Statistical tests show that, except for wheat, the mass functions τ(q) are nonlinear, suggesting the deviation from mono-fractality. However, using iterated amplitude adjusted Fourier transform (IAAFT) time series surrogates, comprehensive statistical tests on the singularity width Δα and spectrum difference Δf show that the time series of GOI, wheat and rice hardly exhibit multifractal nature, the soyabeans index might possess multifractal nature, and the maize and barley time series exhibit intrinsic multifractal behavior since their apparent multifractality cannot be explained by the broad return distribution and linear correlations. Our analysis shows that only part of the grain indices exhibit multifractal nature and calls for further analysis about the complex behaviors of the global grain spot markets by considering the impacts of internal and external shocks.