This is the GitHub repository for the paper "Measuring Inequality Beyond the Gini Coefficient May Clarify Conflicting Findings" by Kristin Blesch, Oliver Hauser & Jon M. Jachimowicz (2022). The paper can be divided into three parts. The respective folders each contain a specific README file that guides through reproducing our results.
Part 1 Lorenz curve estimation using MLE and NLS on a US county and state level. Relevant code can be found in the respective folder including .csv files with estimated Ortega parameters on a US county and state level. You can browse through the interactive visualization of these estimates, or you can have a look at the following Figure for a static preview, where Panel A depicts the Gini coefficient, Panel B Ortega parameter alpha and Panel C Ortega parameter gamma on a US county level.
Part 2 Ortega Lorenz curve simulation to investigate what features about the income distribution each of the parameters capture. We provide relevant code in this repository and an interactive RShiny tool to facilitate understanding the Ortega parameters and enable users to visually compare various Lorenz curves.
Part 3 An exploratory study correlating the Ortega parameters to other county-level characteristics. Relevant code is provided in the respective folder. As a visualization of our main results, consider Figure 5 from the paper:
Figure 5: A two-parameter Ortega approach reveals significant correlations between inequality and policy outcomes that the Gini coefficient misses. Note that the confidence level is 0.9995, using a Bonferroni correction. The figure shows the sub-sample of covariates (33 out of 100) for which the Pearson correlations between county-level variables were not significantly related to the Gini coefficient but exhibited at least one statistically significant (partial) correlation with the Ortega parameters. Abbreviations: M - male; F - female; Q - income quartile; Frac. - fraction; raceadj. - race adjusted