Fluctuation-Dissipation Theory of Input-Output Interindustrial Correlations

TL;DR

This paper applies fluctuation-dissipation theory to Japanese industrial production data, revealing intrinsic business cycles, genuine interindustrial correlations, and external stimuli, even in a far-from-equilibrium economic system.

Contribution

It demonstrates the application of fluctuation-dissipation theory to macroeconomic data, identifying meaningful interindustrial correlations and external influences.

Findings

Two dominant eigenmodes capture intrinsic business cycles.

Correlation matrix describes genuine interindustrial relations.

External stimuli linked to global economic crisis are identified.

Abstract

In this study, the fluctuation-dissipation theory is invoked to shed light on input-output interindustrial relations at a macroscopic level by its application to IIP (indices of industrial production) data for Japan. Statistical noise arising from finiteness of the time series data is carefully removed by making use of the random matrix theory in an eigenvalue analysis of the correlation matrix; as a result, two dominant eigenmodes are detected. Our previous study successfully used these two modes to demonstrate the existence of intrinsic business cycles. Here a correlation matrix constructed from the two modes describes genuine interindustrial correlations in a statistically meaningful way. Further it enables us to quantitatively discuss the relationship between shipments of final demand goods and production of intermediate goods in a linear response framework. We also investigate distinctive external stimuli for the Japanese economy exerted by the current global economic crisis. These stimuli are derived from residuals of moving average fluctuations of the IIP remaining after subtracting the long-period components arising from inherent business cycles. The observation reveals that the fluctuation-dissipation theory is applicable to an economic system that is supposed to be far from physical equilibrium.