Regime-Dependent Topological Properties of Biofuels Networks

We analyze the relationships between biodiesel, ethanol and related fuels and agricultural commodities with a use of minimal spanning trees and hierarchical trees and Granger causality. We construct this trees for different frequencies (weekly, monthly and quarterly). We find that in short-term, both ethanol and biodiesel are very weakly connected with the other commodities. In medium and long term, the network structure becomes more interesting. We especially concentrate on the links and comovements between the commodities for different stages of the market based on the level of prices of important commodities or groups of commodities. Such approach in general confirms the separation of network into fuels and food branches for majority of analyzed cases.