A new three-parameter extension of the inverse Gaussian distribution

Víctor Leiva; Antonio Sanhueza; Andrés Silva; Manuel Galea

doi:10.1016/j.spl.2007.11.020

A new three-parameter extension of the inverse Gaussian distribution

Víctor Leiva, Antonio Sanhueza, Andrés Silva, Manuel Galea

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

9 Scopus citations

Abstract

In this article, we introduce a new model that extends the inverse Gaussian distribution. This model is obtained when a parameter is incorporated into the logarithmic inverse Gaussian distribution producing great flexibility for fitting non-negative data. We present a comprehensive treatment of the properties of this model, including a derivation of the analytical shapes of the density, distribution, and hazard functions, as well as the moments. Furthermore, we illustrate the use of this model by means of an example using likelihood methods. We show that the new model presents an excellent fit for the analyzed data.

Original language	English
Pages (from-to)	1266-1273
Number of pages	8
Journal	Statistics and Probability Letters
Volume	78
Issue number	11
DOIs	https://doi.org/10.1016/j.spl.2007.11.020
State	Published - 15 Aug 2008
Externally published	Yes

Keywords

Birnbaum-Saunders distribution
Hazard function
Likelihood methods
Moments
Sinh-normal distribution

Access to Document

10.1016/j.spl.2007.11.020

Cite this

@article{c3040dda6ec04d6297dfc1cb297e9c48,

title = "A new three-parameter extension of the inverse Gaussian distribution",

abstract = "In this article, we introduce a new model that extends the inverse Gaussian distribution. This model is obtained when a parameter is incorporated into the logarithmic inverse Gaussian distribution producing great flexibility for fitting non-negative data. We present a comprehensive treatment of the properties of this model, including a derivation of the analytical shapes of the density, distribution, and hazard functions, as well as the moments. Furthermore, we illustrate the use of this model by means of an example using likelihood methods. We show that the new model presents an excellent fit for the analyzed data.",

keywords = "Birnbaum-Saunders distribution, Hazard function, Likelihood methods, Moments, Sinh-normal distribution",

author = "V{\'i}ctor Leiva and Antonio Sanhueza and Andr{\'e}s Silva and Manuel Galea",

note = "Funding Information: The authors wish to thank the Editor and referees for their helpful comments that aided in improving this article. This study was supported by the grants: FONDECYT 1080326 and DIUFRO 120621, Chile. ",

year = "2008",

month = aug,

day = "15",

doi = "10.1016/j.spl.2007.11.020",

language = "English",

volume = "78",

pages = "1266--1273",

journal = "Statistics and Probability Letters",

issn = "0167-7152",

publisher = "Elsevier",

number = "11",

}

TY - JOUR

T1 - A new three-parameter extension of the inverse Gaussian distribution

AU - Leiva, Víctor

AU - Sanhueza, Antonio

AU - Silva, Andrés

AU - Galea, Manuel

N1 - Funding Information: The authors wish to thank the Editor and referees for their helpful comments that aided in improving this article. This study was supported by the grants: FONDECYT 1080326 and DIUFRO 120621, Chile.

PY - 2008/8/15

Y1 - 2008/8/15

N2 - In this article, we introduce a new model that extends the inverse Gaussian distribution. This model is obtained when a parameter is incorporated into the logarithmic inverse Gaussian distribution producing great flexibility for fitting non-negative data. We present a comprehensive treatment of the properties of this model, including a derivation of the analytical shapes of the density, distribution, and hazard functions, as well as the moments. Furthermore, we illustrate the use of this model by means of an example using likelihood methods. We show that the new model presents an excellent fit for the analyzed data.

AB - In this article, we introduce a new model that extends the inverse Gaussian distribution. This model is obtained when a parameter is incorporated into the logarithmic inverse Gaussian distribution producing great flexibility for fitting non-negative data. We present a comprehensive treatment of the properties of this model, including a derivation of the analytical shapes of the density, distribution, and hazard functions, as well as the moments. Furthermore, we illustrate the use of this model by means of an example using likelihood methods. We show that the new model presents an excellent fit for the analyzed data.

KW - Birnbaum-Saunders distribution

KW - Hazard function

KW - Likelihood methods

KW - Moments

KW - Sinh-normal distribution

UR - http://www.scopus.com/inward/record.url?scp=47749087539&partnerID=8YFLogxK

U2 - 10.1016/j.spl.2007.11.020

DO - 10.1016/j.spl.2007.11.020

M3 - Article

AN - SCOPUS:47749087539

SN - 0167-7152

VL - 78

SP - 1266

EP - 1273

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

IS - 11

ER -

A new three-parameter extension of the inverse Gaussian distribution

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this