Authors: Cedric Conol, Rosely Peña

Abstract

We develop a view of the global commodity price network by interpreting the commodity price data as a signed network in which commodities act as nodes while the pair-wise correlations of commodities represent links that connect one node from another. We show that it is possible to quantify the complexity of the global commodity price network by characterizing the structure of a weighted signed network. Furthermore, we simulate the possible effects of the pandemic to the network by removing products that are heavily exported by China. Results show that network measures are affected by the market behavior of commodities. Commodities having positive average degree values are usually characterized by high demand due to their uses on different industries while Commodities having negative average degree values are characterized by a price-sensitive market. We also found that commodities that are staple in food processing industries are usually the most influential nodes in the network. These findings are useful for commodity market policymakers as an indicator based on which they can interfere with the markets before the markets make a drastic change, and for trade investors to properly diversify trading portfolios.

Note: If you wish to have a copy of the data and the code used in this project, you may send me a message via Linked in.