Are your data really Pareto distributed?

DOI10.1016/J.PHYSA.2013.07.061zbMATH Open1395.62006arXiv1306.0100OpenAlexW2128442773MaRDI QIDQ1673333FDOQ1673333

Pasquale Cirillo

Publication date: 11 September 2018

Published in: Physica A (Search for Journal in Brave)

Abstract: Pareto distributions, and power laws in general, have demonstrated to be very useful models to describe very different phenomena, from physics to finance. In recent years, the econophysical literature has proposed a large amount of papers and models justifying the presence of power laws in economic data. Most of the times, this Paretianity is inferred from the observation of some plots, such as the Zipf plot and the mean excess plot. If the Zipf plot looks almost linear, then everything is ok and the parameters of the Pareto distribution are estimated. Often with OLS. Unfortunately, as we show in this paper, these heuristic graphical tools are not reliable. To be more exact, we show that only a combination of plots can give some degree of confidence about the real presence of Paretianity in the data. We start by reviewing some of the most important plots, discussing their points of strength and weakness, and then we propose some additional tools that can be used to refine the analysis.

Full work available at URL: https://arxiv.org/abs/1306.0100

Recommendations

• The Pareto, Zipf and other power laws
• A review of goodness of fit tests for Pareto distributions
• The testing of Pareto distributions
• Strong, weak and false inverse power laws
• A unified approach of testing for discrete and continuous Pareto laws

zbMATH Keywords

power lawsZipf plotgraphical toolsmeplotParetianity

Mathematics Subject Classification ID

Statistics of extreme values; tail inference (62G32) Statistical methods; economic indices and measures (91B82) Graphical methods in statistics (62A09)

Cites Work

• Power-Law Distributions in Empirical Data
• Title not available (Why is that?)
• Title not available (Why is that?)
• Statistical Size Distributions in Economics and Actuarial Sciences
• Modeling Income Distributions and Lorenz Curves
• Scale-Free Networks
• A discussion on mean excess plots
• Cross-correlations between volume change and price change
• Laws of Small Numbers: Extremes and Rare Events
• Parametric Lorenz Curves: Models and Applications
• On testing extreme value conditions
• Rank − 1 / 2: A Simple Way to Improve the OLS Estimation of Tail Exponents

Cited In (16)

• On the parametric description of log-growth rates of Romanian city sizes
• Dynamical fitness models: evidence of universality classes for preferential attachment graphs
• A generalization of the power law distribution with nonlinear exponent
• Gini estimation under infinite variance
• On the statistical properties and tail risk of violent conflicts
• On the statistical properties of viral misinformation in online social media
• From concentration profiles to concentration maps. New tools for the study of loss distributions
• The testing of Pareto distributions
• Expected shortfall estimation for apparently infinite-mean models of operational risk
• A robust semi-parametric approach for measuring income inequality in Malaysia
• Graphical representations and associated goodness-of-fit tests for Pareto and log-normal distributions based on inequality curves
• The distribution of strike size: empirical evidence from Europe and north America in the 19th and 20th centuries
• The Pareto, Zipf and other power laws
• Is the wealth of the Forbes 400 lists really Pareto distributed?
• How much better is disaggregate data?
• Five degrees of randomness

Uses Software

• plfit

This page was built for publication: Are your data really Pareto distributed?

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1673333)