Asymptotic distributions for weighted power sums of extreme values

TL;DR

This paper derives the asymptotic distributions of weighted power sums of extreme values from heavy-tailed distributions and introduces new estimators for the tail index parameter.

Contribution

It provides the first asymptotic normality results for weighted power sums of extreme values under heavy-tail conditions, enabling improved tail index estimation.

Findings

Asymptotic normality of weighted power sums established

New class of estimators for tail index γ proposed

Results applicable to heavy-tail models with known constraints

Abstract

Let $X_{1,n}\le\cdots\le X_{n,n}$ be the order statistics of $n$ independent random variables with a common distribution function $F$ having right heavy tail with tail index $\gamma$. Given known constants $d_{i,n}$, $1\le i\le n$, consider the weighted power sums $\sum^{k_n}_{i=1}d_{n+1-i,n}\log^pX_{n+1-i,n}$, where $p>0$ and the $k_n$ are positive integers such that $k_n\to\infty$ and $k_n/n\to0$ as $n\to\infty$. Under some constraints on the weights $d_{i,n}$, we prove asymptotic normality for the power sums over the whole heavy-tail model. We apply the obtained result to construct a new class of estimators for the parameter $\gamma$.