TITLE:
Distributed Trimmed Hill Estimator
AUTHORS:
Tao Guo
KEYWORDS:
Extreme Value Index, Distributed Trimmed Hill Estimator
JOURNAL NAME:
Journal of Applied Mathematics and Physics,
Vol.11 No.12,
December
27,
2023
ABSTRACT: Proceeded from trimmed Hill estimators and distributed inference, a new distributed version of trimmed Hill estimator for heavy tail index is proposed. Considering the case where the number of observations involved in each machine can be either the same or different and either fixed or varying to the total sample size, its consistency and asymptotic normality are discussed. Simulation studies are particularized to show the new estimator performs almost in line with the trimmed Hill estimator.