Trends in Income Distribution in Italy: A Non-Parametric 90 and a Semi-Parametric Analysis

Using kernel density estimation and mixture models, household size-adjusted income distributions in Italy are cross-sectionally examined over the period 1987-2002. A non-parametric test is used to assess the number of modes in the distributions. Evidence shows that income tends to cluster around more than one point. This gives good reason to model the shapes by a finite mixture density with an appropriate choice of components which represent homogeneous sub-populations. Effects of social and demographic factors on the probability of households to belong to one of the components of the mixture are identified by a compositional data analysis.