Fig 6. Distribution of the unambiguous Gini indices of countries of the test data set derived from the information provided by the United Nations University World Institute for Development Economics https://www.wider.unu.edu/download/WIID3.3.

Left: The density distribution is presented as probability density function, estimated by means of the Pareto Density Estimation (PDE [19]; black line). A Gaussian mixture model (GMM) given as $p\left(x\right)={\Sigma }_{i=0}^M{w}_{i}N(x|m_i,s_i)$, where m_i, s_i and w_i are the parameters mean, standard deviation and relative weight of each of the Gaussians, respectively, was fit (red line) to the data, with a number of mixes of M = 4 (Gaussian, G #1 .. G #4) shown as differently colored lines. The Bayesian boundaries between the Gaussians are shown as perpendicular magenta-colored lines. Empty (white) regions indicate either lack of information or information judged as of poor quality by the publisher. The Fig has been created using the R software package (version 3.4.0 for Linux; http://CRAN.R-project.org/ [18]); specifically, the GMM was drawn using our R package “AdaptGauss” (https://cran.r-project.org/package=AdaptGauss [25]). Right: A quantile-quantile plot comparing the observed distribution of standardized Gini indices with the distribution expected from the GMM (abscissa).