Parameter estimation of the Pareto distribution using a pivotal quantity

Joseph H.T. Kim; Sanghyun Ahn; Soohan Ahn

doi:10.1016/j.jkss.2017.01.004

Parameter estimation of the Pareto distribution using a pivotal quantity

Joseph H.T. Kim, Sanghyun Ahn, Soohan Ahn

Department of Applied Statistics

Research output: Contribution to journal › Article › peer-review

11 Citations (Scopus)

Abstract

In estimating the parameters of the two-parameter Pareto distribution it is well known that the performance of the maximum likelihood estimator deteriorates when sample sizes are small or the underlying model is contaminated. In this paper we propose a new parameter estimator that utilizes a pivotal quantity based on the regression framework, allowing separate estimation of the two parameters in a straightforward manner. The consistency of the estimator is also established. Simulation studies show that the proposed estimator is a competitive, well-rounded robust estimator for both Pareto and contaminated Pareto datasets when the sample sizes are small.

Original language	English
Pages (from-to)	438-450
Number of pages	13
Journal	Journal of the Korean Statistical Society
Volume	46
Issue number	3
DOIs	https://doi.org/10.1016/j.jkss.2017.01.004
Publication status	Published - 2017 Sept 1

Bibliographical note

Funding Information:
The research of first author is supported by Basic Science Research Program of the National Research Foundation of Korea ( NRF-2015R1A1A1A05027336 ).

Funding Information:
The work of corresponding author was supported by the 2016 Research Fund of the University of Seoul .

Publisher Copyright:
© 2017 The Korean Statistical Society

All Science Journal Classification (ASJC) codes

Statistics and Probability

Access to Document

10.1016/j.jkss.2017.01.004

Cite this

@article{2a961f69907e498480693309abd2c684,

title = "Parameter estimation of the Pareto distribution using a pivotal quantity",

abstract = "In estimating the parameters of the two-parameter Pareto distribution it is well known that the performance of the maximum likelihood estimator deteriorates when sample sizes are small or the underlying model is contaminated. In this paper we propose a new parameter estimator that utilizes a pivotal quantity based on the regression framework, allowing separate estimation of the two parameters in a straightforward manner. The consistency of the estimator is also established. Simulation studies show that the proposed estimator is a competitive, well-rounded robust estimator for both Pareto and contaminated Pareto datasets when the sample sizes are small.",

author = "Kim, {Joseph H.T.} and Sanghyun Ahn and Soohan Ahn",

note = "Funding Information: The research of first author is supported by Basic Science Research Program of the National Research Foundation of Korea ( NRF-2015R1A1A1A05027336 ). Funding Information: The work of corresponding author was supported by the 2016 Research Fund of the University of Seoul . Publisher Copyright: {\textcopyright} 2017 The Korean Statistical Society",

year = "2017",

month = sep,

day = "1",

doi = "10.1016/j.jkss.2017.01.004",

language = "English",

volume = "46",

pages = "438--450",

journal = "Journal of the Korean Statistical Society",

issn = "1226-3192",

publisher = "Korean Statistical Society",

number = "3",

}

TY - JOUR

T1 - Parameter estimation of the Pareto distribution using a pivotal quantity

AU - Kim, Joseph H.T.

AU - Ahn, Sanghyun

AU - Ahn, Soohan

N1 - Funding Information: The research of first author is supported by Basic Science Research Program of the National Research Foundation of Korea ( NRF-2015R1A1A1A05027336 ). Funding Information: The work of corresponding author was supported by the 2016 Research Fund of the University of Seoul . Publisher Copyright: © 2017 The Korean Statistical Society

PY - 2017/9/1

Y1 - 2017/9/1

N2 - In estimating the parameters of the two-parameter Pareto distribution it is well known that the performance of the maximum likelihood estimator deteriorates when sample sizes are small or the underlying model is contaminated. In this paper we propose a new parameter estimator that utilizes a pivotal quantity based on the regression framework, allowing separate estimation of the two parameters in a straightforward manner. The consistency of the estimator is also established. Simulation studies show that the proposed estimator is a competitive, well-rounded robust estimator for both Pareto and contaminated Pareto datasets when the sample sizes are small.

AB - In estimating the parameters of the two-parameter Pareto distribution it is well known that the performance of the maximum likelihood estimator deteriorates when sample sizes are small or the underlying model is contaminated. In this paper we propose a new parameter estimator that utilizes a pivotal quantity based on the regression framework, allowing separate estimation of the two parameters in a straightforward manner. The consistency of the estimator is also established. Simulation studies show that the proposed estimator is a competitive, well-rounded robust estimator for both Pareto and contaminated Pareto datasets when the sample sizes are small.

UR - http://www.scopus.com/inward/record.url?scp=85011547328&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85011547328&partnerID=8YFLogxK

U2 - 10.1016/j.jkss.2017.01.004

DO - 10.1016/j.jkss.2017.01.004

M3 - Article

AN - SCOPUS:85011547328

SN - 1226-3192

VL - 46

SP - 438

EP - 450

JO - Journal of the Korean Statistical Society

JF - Journal of the Korean Statistical Society

IS - 3

ER -

Parameter estimation of the Pareto distribution using a pivotal quantity

Abstract

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this