Modelling income, wealth, and expenditure data by use of Econophysics

TL;DR

This paper explores the application of physics-inspired distributions like logistic and Fermi-Dirac to model income, wealth, and expenditure data, proposing a dynamic methodology and a Hamiltonian-based utility model, across diverse countries.

Contribution

It introduces a novel approach combining physics distributions with economic data, including a dynamic methodology and a Hamiltonian utility model for macroeconomic analysis.

Findings

Logistic distribution fits well across entire income ranges.

Fermi-Dirac distribution reveals correlations with macroeconomic variables.

Polynomial distributions effectively model income data.

Abstract

In the present paper, we identify several distributions from Physics and study their applicability to phenomena such as distribution of income, wealth, and expenditure. Firstly, we apply logistic distribution to these data and we find that it fits very well the annual data for the entire income interval including for upper income segment of population. Secondly, we apply Fermi-Dirac distribution to these data. We seek to explain possible correlations and analogies between economic systems and statistical thermodynamics systems. We try to explain their behavior and properties when we correlate physical variables with macroeconomic aggregates and indicators. Then we draw some analogies between parameters of the Fermi-Dirac distribution and macroeconomic variables. Thirdly, as complex systems are modeled using polynomial distributions, we apply polynomials to the annual sets of data and we find that it fits very well also the entire income interval. Fourthly, we develop a new methodology to approach dynamically the income, wealth, and expenditure distribution similarly with dynamical complex systems. This methodology was applied to different time intervals consisting of consecutive years up to 35 years. Finally, we develop a mathematical model based on a Hamiltonian that maximizes utility function applied to Ramsey model using Fermi-Dirac and polynomial utility functions. We find some theoretical connections with time preference theory. We apply these distributions to a large pool of data from countries with different levels of development, using different methods for calculation of income, wealth, and expenditure.