Google matrix analysis of the multiproduct world trade network

LEONARDO ERMANN; DIMA L. SHEPELYANSKY

EUROPEAN PHYSICAL JOURNAL B - CONDENSED MATTER

SPRINGER

Lugar: Berlin; Año: 2015 vol. 88 p. 84 - 103

1434-6028

Using the United Nations COMTRADE database [United Nations Commodity Trade StatisticsDatabase, available at: http://comtrade.un.org/db/. Accessed November (2014)] we construct the Googlematrix G of multiproduct world trade between the UN countries and analyze the properties of trade flows onthis network for years 1962−2010. This construction, based on Markov chains, treats all countries on equaldemocratic grounds independently of their richness and at the same time it considers the contributionsof trade products proportionally to their trade volume. We consider the trade with 61 products for upto 227 countries. The obtained results show that the trade contribution of products is asymmetric: someof them are export oriented while others are import oriented even if the ranking by their trade volumeis symmetric in respect to export and import after averaging over all world countries. The constructionof the Google matrix allows to investigate the sensitivity of trade balance in respect to price variationsof products, e.g. petroleum and gas, taking into account the world connectivity of trade links. The tradebalance based on PageRank and CheiRank probabilities highlights the leading role of China and otherBRICS countries in the world trade in recent years. We also show that the eigenstates of G with largeeigenvalues select specific trade communities.