International Trade: a Reinforced Urn Network Model

TL;DR

This paper introduces a novel reinforced urn network model that explains key empirical features of the international trade network, including power law distributions and clustering, using a unified probabilistic framework.

Contribution

It develops a theoretical model linking Polya urns and reinforced processes to international trade, enabling likelihood-based estimation and capturing complex network features.

Findings

Model reproduces power law tail behavior of trade flows.

Empirical analysis confirms the model's ability to replicate trade network regularities.

Likelihood estimation is feasible and effective in real-world data.

Abstract

We propose a unified modelling framework that theoretically justifies the main empirical regularities characterizing the international trade network. Each country is associated to a Polya urn whose composition controls the propensity of the country to trade with other countries. The urn composition is updated through the walk of the Reinforced Urn Process of Muliere et al. (2000). The model implies a local preferential attachment scheme and a power law right tail behaviour of bilateral trade flows. Different assumptions on the urns' reinforcement parameters account for local clustering, path-shortening and sparsity. Likelihood-based estimation approaches are facilitated by feasible likelihood analytical derivation in various network settings. A simulated example and the empirical results on the international trade network are discussed.