Authors: Periklis Gogas, Theophilos Papadimitriou, Georgios Sarantitis

Title: Network Analysis for Central Banks

Abstract

Graph Theory and network analysis have recently been incorporated in the financial analysis’ departments of Central Banks. In this study we introduce the utility of complex networks as a tool for macroeconomic modelling and decision making. More specifically we use simple network metrics, community detection algorithms and an optimization technique called the Threshold-Minimum Dominating Set to provide a multi-level analysis of the United Kingdom’s Consumer Price Index (CPI). Our empirical analysis yields various interesting results. The 85 classes that comprise the UK CPI display complex inter-relations that cannot be captured by classic econometric analysis. UK’s CPI presents specific behavior prior to a crisis that could be exploited to produce an alert gauge for future shocks. A small group of those classes can be used to construct a new index that satisfies the criteria and can be used as an alternative measure of core inflation by the Bank of England.