International trade of agricultural products as analyzed by complex network method

The research paradigm of complex network based on statistical physics and graph theory has sprung up in the last decade to provide a new global perspective for international trade, especially for trade of agricultural commodities. We in the present work apply the theories of complex network to investigate positions of
different countries in international trade of agricultural commodities. Countries are classified into three categories based on “core/periphery” structure using Weighted Extremal Optimisation algorithm and Coarse Graining process. Countries are ranked with the aid of network node centralities, which presents world
agricultural commodity trade as a closed, imbalanced, diversified and multi-polar development. Improved bootstrap percolation was introduced to simulate cascading influences following breaking down of bilateral agricultural commodity trade relations.