Detecting spatial homogeneity in the world trade web with Detrended Fluctuation Analysis

TL;DR

This paper introduces a novel application of Detrended Fluctuation Analysis to assess the impact of spatial distances on the World Trade Web, revealing a declining influence of distances over time.

Contribution

It extends DFA methodology to spatially ordered data and applies it to the WTW, challenging gravity model predictions about distance effects.

Findings

Distances' influence on trade relationships declines over time.

DFA reveals autocorrelations in spatial trade data.

Spatial homogeneity increases in the WTW.

Abstract

In a spatially embedded network, that is a network where nodes can be uniquely determined in a system of coordinates, links' weights might be affected by metric distances coupling every pair of nodes (dyads). In order to assess to what extent metric distances affect relationships (link's weights) in a spatially embedded network, we propose a methodology based on DFA (Detrended Fluctuation Analysis). DFA is a well developed methodology to evaluate autocorrelations and estimate long-range behaviour in time series. We argue it can be further extended to spatially ordered series in order to assess autocorrelations in values. A scaling exponent of 0.5 (uncorrelated data) would thereby signal a perfect homogeneous space embedding the network. We apply the proposed methodology to the World Trade Web (WTW) during the years 1949-2000 and we find, in some contrast with predictions of gravity models, a declining influence of distances on trading relationships.