Analysis of the Gini Coefficient: evidence for the state of Puebla with a non-linear approach

Abstract

The objective of this article is to measure the Gini Coefficient by Nonlinear Ordinary Least Squares (NLOLS), following a different methodology for the design of the Lorenz Curve so that it fits better for the purposes of estimation. The estimation is based on the fact that the Lorenz curve can be simulated through a function of root n with coefficients a and b of the form , where and the Gini Coefficient can be calculated in the form Gini = 1-b. The main findings using data from the EnBienComún survey of the State of Puebla 2015, are that the estimate of the inequality is greater than that reported by CONEVAL in 2015 for the State of Puebla, this estimate is statistically significant. AIC and BIC criteria are favorable for the specification of the estimate in terms of sample size