Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Klar, Bernhard
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2310.01076
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916192846872576
author Klar, Bernhard
author_facet Klar, Bernhard
contents We propose a mean functional which exists for any probability distributions, and which characterizes the Pareto distribution within the set of distributions with finite left endpoint. This is in sharp contrast to the mean excess plot which is not meaningful for distributions without existing mean, and which has a nonstandard behaviour if the mean is finite, but the second moment does not exist. The construction of the plot is based on the so called principle of a single huge jump, which differentiates between distributions with moderately heavy and super heavy tails. We present an estimator of the tail function based on $U$-statistics and study its large sample properties. The use of the new plot is illustrated by several loss datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2310_01076
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A Pareto tail plot without moment restrictions
Klar, Bernhard
Methodology
We propose a mean functional which exists for any probability distributions, and which characterizes the Pareto distribution within the set of distributions with finite left endpoint. This is in sharp contrast to the mean excess plot which is not meaningful for distributions without existing mean, and which has a nonstandard behaviour if the mean is finite, but the second moment does not exist. The construction of the plot is based on the so called principle of a single huge jump, which differentiates between distributions with moderately heavy and super heavy tails. We present an estimator of the tail function based on $U$-statistics and study its large sample properties. The use of the new plot is illustrated by several loss datasets.
title A Pareto tail plot without moment restrictions
topic Methodology
url https://arxiv.org/abs/2310.01076