Properties of Selected Inequality Measures Based on Quantiles and Their Application to the Analysis of Income Distribution in Poland by Macroregion

Abstract

Quantiles of income distributions are often applied to the estimation of various inequality, poverty and wealth characteristics. They are traditionally estimated using the classical quantile estimator based on a relevant order statistic. The main objective of the paper is to compare the classical, Huang-Brill and Bernstein estimators for these measures from the point of view of their statistical properties. Several Monte Carlo experiments were conducted to assess biases and mean squared errors of income distribution characteristics for different sample sizes under the lognormal or Dagum type-I models. The results of these experiments are used to estimate inequality, poverty and wealth measures in Poland by macroregion on the basis of micro data originating from the Household Budget Survey 2014.