On parameter orthogonality in symmetric and skew models

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Orthogonal and partly orthogonal reparametrisations are provided for certain wide and important families of univariate continuous distributions. First, the orthogonality of parameters in location-scale symmetric families is extended to symmetric distributions involving a third parameter. This sets the scene for consideration of the four-parameter situation in which skewness is also allowed. It turns out that one specific approach to generating such four-parameter families, that of two-piece distributions with a certain parametrisation restriction, has some attractive features with regard to parameter orthogonality which, to the best of our knowledge, are not shared with other four-parameter distributions. Our work also affords partly orthogonal parametrisations of three-parameter two-piece models.
Original languageEnglish
Pages (from-to)758 - 770
Number of pages13
JournalJournal of Statistical Planning and Inference
Volume141
Issue number2
Early online date6 Aug 2010
DOIs
Publication statusPublished - Feb 2011

Fingerprint

Orthogonality
Skew
Parametrization
Model
Reparametrization
Symmetric Distributions
Continuous Distributions
Skewness
Univariate
Statistical Models
Two Parameters
Restriction
Family

Keywords

  • Two-piece distribution

Cite this

On parameter orthogonality in symmetric and skew models. / Jones, M.C.; Anaya-Izquierdo, Karim.

In: Journal of Statistical Planning and Inference, Vol. 141, No. 2, 02.2011, p. 758 - 770.

Research output: Contribution to journalArticle

@article{18a6fafd1b634df78ce91ffa2d766dfa,
title = "On parameter orthogonality in symmetric and skew models",
abstract = "Orthogonal and partly orthogonal reparametrisations are provided for certain wide and important families of univariate continuous distributions. First, the orthogonality of parameters in location-scale symmetric families is extended to symmetric distributions involving a third parameter. This sets the scene for consideration of the four-parameter situation in which skewness is also allowed. It turns out that one specific approach to generating such four-parameter families, that of two-piece distributions with a certain parametrisation restriction, has some attractive features with regard to parameter orthogonality which, to the best of our knowledge, are not shared with other four-parameter distributions. Our work also affords partly orthogonal parametrisations of three-parameter two-piece models.",
keywords = "Two-piece distribution",
author = "M.C. Jones and Karim Anaya-Izquierdo",
year = "2011",
month = "2",
doi = "10.1016/j.jspi.2010.07.021",
language = "English",
volume = "141",
pages = "758 -- 770",
journal = "Journal of Statistical Planning and Inference",
issn = "0378-3758",
publisher = "Elsevier",
number = "2",

}

TY - JOUR

T1 - On parameter orthogonality in symmetric and skew models

AU - Jones, M.C.

AU - Anaya-Izquierdo, Karim

PY - 2011/2

Y1 - 2011/2

N2 - Orthogonal and partly orthogonal reparametrisations are provided for certain wide and important families of univariate continuous distributions. First, the orthogonality of parameters in location-scale symmetric families is extended to symmetric distributions involving a third parameter. This sets the scene for consideration of the four-parameter situation in which skewness is also allowed. It turns out that one specific approach to generating such four-parameter families, that of two-piece distributions with a certain parametrisation restriction, has some attractive features with regard to parameter orthogonality which, to the best of our knowledge, are not shared with other four-parameter distributions. Our work also affords partly orthogonal parametrisations of three-parameter two-piece models.

AB - Orthogonal and partly orthogonal reparametrisations are provided for certain wide and important families of univariate continuous distributions. First, the orthogonality of parameters in location-scale symmetric families is extended to symmetric distributions involving a third parameter. This sets the scene for consideration of the four-parameter situation in which skewness is also allowed. It turns out that one specific approach to generating such four-parameter families, that of two-piece distributions with a certain parametrisation restriction, has some attractive features with regard to parameter orthogonality which, to the best of our knowledge, are not shared with other four-parameter distributions. Our work also affords partly orthogonal parametrisations of three-parameter two-piece models.

KW - Two-piece distribution

UR - http://www.scopus.com/inward/record.url?scp=77955526496&partnerID=8YFLogxK

UR - http://www.sciencedirect.com/science/article/pii/S037837581000368X

U2 - 10.1016/j.jspi.2010.07.021

DO - 10.1016/j.jspi.2010.07.021

M3 - Article

VL - 141

SP - 758

EP - 770

JO - Journal of Statistical Planning and Inference

JF - Journal of Statistical Planning and Inference

SN - 0378-3758

IS - 2

ER -