Cluster structure of EU-15 countries derived from the correlation matrix analysis of macroeconomic index fluctuations

Abstract

The statistical distances between countries, calculated for various moving average time windows, are mapped into the ultrametric subdominant space as in classical Minimal Spanning Tree methods. The Moving Average Minimal Length Path (MAMLP) algorithm allows a decoupling of fluctuations with respect to the mass center of the system from the movement of the mass center itself. A Hamiltonian representation given by a factor graph is used and plays the role of cost function. The present analysis pertains to 11 macroeconomic (ME) indicators, namely the GDP (x1), Final Consumption Expenditure (x2), Gross Capital Formation (x3), Net Exports (x4), Consumer Price Index (y1), Rates of Interest of the Central Banks (y2), Labour Force (z1), Unemployment (z2), GDP/hour worked (z3), GDP/capita (w1) and Gini coefficient (w2). The target group of countries is composed of 15 EU countries, data taken between 1995 and 2004. By two different methods (the Bipartite Factor Graph Analysis and the Correlation Matrix Eigensystem Analysis) it is found that the strongly correlated countries with respect to the macroeconomic indicators fluctuations can be partitioned into stable clusters.

Cited by in corpus (7)

• Will the US Economy Recover in 2010? A Minimal Spanning Tree Study
• Clusters in weighted macroeconomic networks : the EU case. Introducing the overlapping index of GDP/capita fluctuation correlations
• Exploring how innovation strategies at time of crisis influence performance: a cluster analysis perspective
• Dynamic Peer-to-Peer Competition
• Organization of networks with tagged nodes and biased links: a priori distinct communities. The case of Intelligent Design Proponents and Darwinian Evolution Defenders
• Econophysics: Comments on a few Applications, Successes, Methods, & Models
• On religion and language evolutions seen through mathematical and agent based models