Gender wage inequality: new evidence from penalized expectile regression

Abstract The Machado-Mata decomposition building on quantile regression has been extensively analyzed in the literature focusing gender wage inequality. In this study, we generalize to expectile framework, which, best of our knowledge, never applied strand literature. contrast, recent years, expectiles have gained increasing attention other contexts as an alternative traditional quantiles, providing useful statistical and computational properties. We flexibly deal with high-dimensional problems by employing Least Absolute Shrinkage Selection Operator. empirical analysis focuses pay gap Germany Italy. find that depending estimation approach (i.e. or regression) results substantially differ along some regions distribution, whereas they are similar for others. From a policy perspective, finding is important it affects conclusions about glass ceiling sticky floors.