Reduced bias Hill estimators

Ivanilda Cabral; Frederico Caeiro; M. Ivette Gomes

doi:10.1063/1.4968687

Reduced bias Hill estimators

Ivanilda Cabral, Frederico Caeiro, M. Ivette Gomes

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Citations (Scopus)

Abstract

For heavy tails, classical extreme value index estimators, like the Hill estimator, are usually asymptotically biased. Consequently those estimators are quite sensitive to the number of top order statistics used in the estimation. The recent minimum-variance reduced-bias extreme value index estimators enable us to remove the dominant component of asymptotic bias and keep the asymptotic variance of the new estimators equal to the asymptotic variance of the Hill estimator. In this paper a new minimum-variance reduced-bias extreme value index estimator is introduced, and its non degenerate asymptotic behaviour is studied. A comparison with another important minimum-variance reduced-bias extreme value index estimator is also provided.

Original language	English
Title of host publication	International Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016
Publisher	AIP - American Institute of Physics
Volume	1790
ISBN (Electronic)	978-0-7354-1454-9
DOIs	https://doi.org/10.1063/1.4968687
Publication status	Published - 6 Dec 2016
Event	International Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016 - Athens, Greece Duration: 17 Mar 2016 → 20 Mar 2016

Conference

Conference	International Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016
Country/Territory	Greece
City	Athens
Period	17/03/16 → 20/03/16

Keywords

Computer Science
Physics

Access to Document

10.1063/1.4968687Licence: Unspecified

Cite this

@inproceedings{b8ce22fa47d743a5b92dff0cfa71ec3b,

title = "Reduced bias Hill estimators",

abstract = "For heavy tails, classical extreme value index estimators, like the Hill estimator, are usually asymptotically biased. Consequently those estimators are quite sensitive to the number of top order statistics used in the estimation. The recent minimum-variance reduced-bias extreme value index estimators enable us to remove the dominant component of asymptotic bias and keep the asymptotic variance of the new estimators equal to the asymptotic variance of the Hill estimator. In this paper a new minimum-variance reduced-bias extreme value index estimator is introduced, and its non degenerate asymptotic behaviour is studied. A comparison with another important minimum-variance reduced-bias extreme value index estimator is also provided.",

keywords = "Computer Science, Physics",

author = "Ivanilda Cabral and Frederico Caeiro and Gomes, {M. Ivette}",

note = "Sem PDF conforme despacho; International Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016 ; Conference date: 17-03-2016 Through 20-03-2016",

year = "2016",

month = dec,

day = "6",

doi = "10.1063/1.4968687",

language = "English",

volume = "1790",

booktitle = "International Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016",

publisher = "AIP - American Institute of Physics",

address = "United States",

}

Cabral, I, Caeiro, F & Gomes, MI 2016, Reduced bias Hill estimators. in International Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016. vol. 1790, 080006, AIP - American Institute of Physics, International Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016, Athens, Greece, 17/03/16. https://doi.org/10.1063/1.4968687

TY - GEN

T1 - Reduced bias Hill estimators

AU - Cabral, Ivanilda

AU - Caeiro, Frederico

AU - Gomes, M. Ivette

N1 - Sem PDF conforme despacho

PY - 2016/12/6

Y1 - 2016/12/6

N2 - For heavy tails, classical extreme value index estimators, like the Hill estimator, are usually asymptotically biased. Consequently those estimators are quite sensitive to the number of top order statistics used in the estimation. The recent minimum-variance reduced-bias extreme value index estimators enable us to remove the dominant component of asymptotic bias and keep the asymptotic variance of the new estimators equal to the asymptotic variance of the Hill estimator. In this paper a new minimum-variance reduced-bias extreme value index estimator is introduced, and its non degenerate asymptotic behaviour is studied. A comparison with another important minimum-variance reduced-bias extreme value index estimator is also provided.

AB - For heavy tails, classical extreme value index estimators, like the Hill estimator, are usually asymptotically biased. Consequently those estimators are quite sensitive to the number of top order statistics used in the estimation. The recent minimum-variance reduced-bias extreme value index estimators enable us to remove the dominant component of asymptotic bias and keep the asymptotic variance of the new estimators equal to the asymptotic variance of the Hill estimator. In this paper a new minimum-variance reduced-bias extreme value index estimator is introduced, and its non degenerate asymptotic behaviour is studied. A comparison with another important minimum-variance reduced-bias extreme value index estimator is also provided.

KW - Computer Science

KW - Physics

UR - http://www.scopus.com/inward/record.url?scp=85008697798&partnerID=8YFLogxK

U2 - 10.1063/1.4968687

DO - 10.1063/1.4968687

M3 - Conference contribution

AN - SCOPUS:85008697798

VL - 1790

BT - International Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016

PB - AIP - American Institute of Physics

T2 - International Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016

Y2 - 17 March 2016 through 20 March 2016

ER -

Reduced bias Hill estimators

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this