A new regression-based tail index estimator

João Nicolau, Paulo M. M. Rodrigues

Research output: Contribution to journalArticle

Abstract

A new regression-based approach for the estimation of the tail
index of heavy-tailed distributions with several important properties is introduced.
First, it provides a bias reduction when compared to available
regression-based methods; second, it is resilient to the choice of the tail
length used for the estimation of the tail index; third, when the effect of
the slowly varying function at infinity of the Pareto distribution vanishes
slowly, it continues to perform satisfactorily; and fourth, it performs well
under dependence of unknown form. An approach to compute the asymptotic
variance under time dependence and conditional heteroskcedasticity
is also provided.
Original languageEnglish
Pages (from-to)667-680
JournalReview of Economics and Statistics
Volume101
Issue number4
DOIs
Publication statusPublished - 1 Oct 2019

Fingerprint

regression
trend
Estimator
Tail index
time
Pareto distribution
Time dependence
Heavy-tailed distribution
Bias reduction

Cite this

@article{b73a9e446a6e4514929bff748a152745,
title = "A new regression-based tail index estimator",
abstract = "A new regression-based approach for the estimation of the tailindex of heavy-tailed distributions with several important properties is introduced.First, it provides a bias reduction when compared to availableregression-based methods; second, it is resilient to the choice of the taillength used for the estimation of the tail index; third, when the effect ofthe slowly varying function at infinity of the Pareto distribution vanishesslowly, it continues to perform satisfactorily; and fourth, it performs wellunder dependence of unknown form. An approach to compute the asymptoticvariance under time dependence and conditional heteroskcedasticityis also provided.",
author = "Jo{\~a}o Nicolau and Rodrigues, {Paulo M. M.}",
year = "2019",
month = "10",
day = "1",
doi = "10.1162/rest_a_00768",
language = "English",
volume = "101",
pages = "667--680",
journal = "Review of Economics and Statistics",
issn = "0034-6535",
publisher = "Elsevier Science B.V., Amsterdam.",
number = "4",

}

A new regression-based tail index estimator. / Nicolau, João; Rodrigues, Paulo M. M.

In: Review of Economics and Statistics, Vol. 101, No. 4, 01.10.2019, p. 667-680.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A new regression-based tail index estimator

AU - Nicolau, João

AU - Rodrigues, Paulo M. M.

PY - 2019/10/1

Y1 - 2019/10/1

N2 - A new regression-based approach for the estimation of the tailindex of heavy-tailed distributions with several important properties is introduced.First, it provides a bias reduction when compared to availableregression-based methods; second, it is resilient to the choice of the taillength used for the estimation of the tail index; third, when the effect ofthe slowly varying function at infinity of the Pareto distribution vanishesslowly, it continues to perform satisfactorily; and fourth, it performs wellunder dependence of unknown form. An approach to compute the asymptoticvariance under time dependence and conditional heteroskcedasticityis also provided.

AB - A new regression-based approach for the estimation of the tailindex of heavy-tailed distributions with several important properties is introduced.First, it provides a bias reduction when compared to availableregression-based methods; second, it is resilient to the choice of the taillength used for the estimation of the tail index; third, when the effect ofthe slowly varying function at infinity of the Pareto distribution vanishesslowly, it continues to perform satisfactorily; and fourth, it performs wellunder dependence of unknown form. An approach to compute the asymptoticvariance under time dependence and conditional heteroskcedasticityis also provided.

U2 - 10.1162/rest_a_00768

DO - 10.1162/rest_a_00768

M3 - Article

VL - 101

SP - 667

EP - 680

JO - Review of Economics and Statistics

JF - Review of Economics and Statistics

SN - 0034-6535

IS - 4

ER -