Resumen de Modeling Mothers' yearly earnings after returning from maternity leave with a bayesian distributional regression model

Tim Föckersperger, Helga Wagner

• The goal of this paper is to estimate and compare the e ects of potential covariates on mothers' yearly earnings in the rst and the sixth year after the maternity leave period in Austria. To go beyond modeling only the mean yearly salary as a function of covariates, we analyze the data with a Bayesian distributional regression model where the response variable is assumed to follow a zero-adjusted Gamma distribution. This allows to model the proportion of mothers with zero income as well as the Gamma-speci c parameters in terms of covariates.