On the penultimate tail behavior of Weibull-type models
Marta Ferreira

TL;DR
This paper investigates the detailed tail behaviors of Weibull-type distributions, revealing a transition in penultimate tail behavior depending on the Weibull-tail coefficient, which refines understanding of their asymptotic properties.
Contribution
It characterizes the penultimate tail behavior of Weibull-type models, showing a phase transition based on the Weibull-tail coefficient , which was not previously detailed.
Findings
Weibull-type distributions exhibit Frchet penultimate tail behavior if >1.
They show Weibull penultimate tail behavior if <1.
The Weibull-tail coefficient determines the tail decay rate.
Abstract
The Gumbel max-domain of attraction corresponds to a null tail index which do not distinguish the different tail weights that might exist between distributions within this class. The Weibull-type distributions form an important subgroup of this latter and includes the so-called \emph{Weibull-tail coefficient}, usually denoted \theta, that specifies the tail behavior, with larger values indicating slower tail decay. Here we shall see that the Weibull-type distributions present a penultimate tail behavior Fr\'echet if \theta>1 and a penultimate tail behavior Weibull whenever \theta<1.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsFinancial Risk and Volatility Modeling · Statistical Distribution Estimation and Applications · Complex Systems and Time Series Analysis
