Inference for types and structured families of commutative orthogonal block structures

Francisco Carvalho, João T. Mexia, Carla Santos, Célia Nunes

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Models with commutative orthogonal block structure, COBS, have orthogonal block structure, OBS, and their least square estimators for estimable vectors are, as it will be shown, best linear unbiased estimator, BLUE. Commutative Jordan algebras will be used to study the algebraic structure of the models and to define special types of models for which explicit expressions for the estimation of variance components are obtained. Once normality is assumed, inference using pivot variables is quite straightforward. To illustrate this class of models we will present unbalanced examples before considering families of models. When the models in a family correspond to the treatments of a base design, the family is structured. It will be shown how, under quite general conditions, the action of the factors in the base design on estimable vectors, can be studied.

Original languageEnglish
Pages (from-to)337-372
Number of pages36
JournalMetrika
Volume78
Issue number3
DOIs
Publication statusPublished - 2015

Fingerprint

Block Structure
Model
Best Linear Unbiased Estimator
Components of Variance
Jordan Algebra
Pivot
Commutative Algebra
Least Squares Estimator
Algebraic Structure
Normality
Family
Inference

Keywords

  • Commutative Jordan algebras
  • Commutative orthogonal block structure
  • Estimation
  • Mixed linear models

Cite this

Carvalho, Francisco ; Mexia, João T. ; Santos, Carla ; Nunes, Célia. / Inference for types and structured families of commutative orthogonal block structures. In: Metrika. 2015 ; Vol. 78, No. 3. pp. 337-372.
@article{fc5b1f05ec424c548f367b2b8b09de4c,
title = "Inference for types and structured families of commutative orthogonal block structures",
abstract = "Models with commutative orthogonal block structure, COBS, have orthogonal block structure, OBS, and their least square estimators for estimable vectors are, as it will be shown, best linear unbiased estimator, BLUE. Commutative Jordan algebras will be used to study the algebraic structure of the models and to define special types of models for which explicit expressions for the estimation of variance components are obtained. Once normality is assumed, inference using pivot variables is quite straightforward. To illustrate this class of models we will present unbalanced examples before considering families of models. When the models in a family correspond to the treatments of a base design, the family is structured. It will be shown how, under quite general conditions, the action of the factors in the base design on estimable vectors, can be studied.",
keywords = "Commutative Jordan algebras, Commutative orthogonal block structure, Estimation, Mixed linear models",
author = "Francisco Carvalho and Mexia, {Jo{\~a}o T.} and Carla Santos and C{\'e}lia Nunes",
note = "Sem PDF",
year = "2015",
doi = "10.1007/s00184-014-0506-8",
language = "English",
volume = "78",
pages = "337--372",
journal = "Metrika",
issn = "0026-1335",
publisher = "Springer Science Business Media",
number = "3",

}

Inference for types and structured families of commutative orthogonal block structures. / Carvalho, Francisco; Mexia, João T.; Santos, Carla; Nunes, Célia.

In: Metrika, Vol. 78, No. 3, 2015, p. 337-372.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Inference for types and structured families of commutative orthogonal block structures

AU - Carvalho, Francisco

AU - Mexia, João T.

AU - Santos, Carla

AU - Nunes, Célia

N1 - Sem PDF

PY - 2015

Y1 - 2015

N2 - Models with commutative orthogonal block structure, COBS, have orthogonal block structure, OBS, and their least square estimators for estimable vectors are, as it will be shown, best linear unbiased estimator, BLUE. Commutative Jordan algebras will be used to study the algebraic structure of the models and to define special types of models for which explicit expressions for the estimation of variance components are obtained. Once normality is assumed, inference using pivot variables is quite straightforward. To illustrate this class of models we will present unbalanced examples before considering families of models. When the models in a family correspond to the treatments of a base design, the family is structured. It will be shown how, under quite general conditions, the action of the factors in the base design on estimable vectors, can be studied.

AB - Models with commutative orthogonal block structure, COBS, have orthogonal block structure, OBS, and their least square estimators for estimable vectors are, as it will be shown, best linear unbiased estimator, BLUE. Commutative Jordan algebras will be used to study the algebraic structure of the models and to define special types of models for which explicit expressions for the estimation of variance components are obtained. Once normality is assumed, inference using pivot variables is quite straightforward. To illustrate this class of models we will present unbalanced examples before considering families of models. When the models in a family correspond to the treatments of a base design, the family is structured. It will be shown how, under quite general conditions, the action of the factors in the base design on estimable vectors, can be studied.

KW - Commutative Jordan algebras

KW - Commutative orthogonal block structure

KW - Estimation

KW - Mixed linear models

UR - http://www.scopus.com/inward/record.url?scp=84939896146&partnerID=8YFLogxK

U2 - 10.1007/s00184-014-0506-8

DO - 10.1007/s00184-014-0506-8

M3 - Article

VL - 78

SP - 337

EP - 372

JO - Metrika

JF - Metrika

SN - 0026-1335

IS - 3

ER -