Adaptive probability weighted moments estimation

Abstract

In this paper, we make use of probability weighted moments of largest observations, in order to build classes of estimators of the extreme value index, the primary parameter in statistics of extremes. Due to the speci city of the estimators, we propose the use of bootstrap computer intensive methods for an adaptive choice of the optimal number of order statistics to be used in the estimation. The developed methodology is applied to a data set in the eld of insurance.