Title: Deriving Lehmer and Hölder means as maximum weighted likelihood estimates for the multivariate exponential family

Abstract: The links between the mean families of Lehmer and H\"older and the weighted maximum likelihood estimator have recently been established in the case of a regular univariate exponential family. In this article, we will extend the outcomes obtained to the multivariate case. This extension provides a probabilistic interpretation of these families of means and could therefore broaden their uses in various applications.