A Density-Based estimator of core/periphery network structures

Abstract

The core/periphery model is increasingly being used to analyze interbank networks, as it is consistent with theoretical models of interbank market structures. We show that existing core/periphery estimators are inaccurate when the network is either highly connected relative to the true proportion of banks in the core, or relatively sparsely connected. We derive a ‘density‑based’ estimator that is designed to overcome these inaccuracies, and show that our estimator outperforms the commonly used estimators in simulations. These results have broad applicability to the network analysis literature. We then use our density‑based estimator to analyze the Australian overnight interbank market, focusing on changes that occurred during the 2007–08 financial crisis.