SENBILIBILIDAD DE LA EVOLUCIÓN DE LA DESIGUALDAD A LAS TÉCNICAS DE INFERENCIA UTILIZADAS. Una aplicación para el índice de Gini en el caso español (1993-2000)

The conclusions regarding the evolution of inequality in countries or regions are frequently deduced from variations in the point estimations of the usual indicators, without considering the impact of sampling errors on the statistical significance of the changes detected. In this paper, we analyze the influence of different inference methods (asymptotic and bootstrap) on the Gini coefficient in detecting the inter-annual variations of this indicator. With this aim, we first carry out a simulation experiment in order to evaluate the behaviour of different methodologies once we know the real value of the population parameters. Secondly, we present a sensitivity study, applied to data in the Spanish case, in order to analyze the influence of the inference technique on the definition of trends in the behaviour of the Gini coefficient, using data from the European Community Household Panel for the period 1993-2000. [ABSTRACT FROM AUTHOR]