Bias-reduced estimation of mean absolute deviation around the median

Abstract

A bias-reduced estimator is proposed for the mean absolute deviation parameter of a median regression model. A workaround is devised for the lack of smoothness in the sense conventionally required in general bias-reduced estimation. A local asymptotic normality property and a Bahadur--Kiefer representation suffice in proving the validity of the bias correction. The proposal is developed under a classical asymptotic regime but, based on simulations, it seems to work also in high-dimensional settings.