Average information REML: An efficient algorithm for variance parameter estimation in linear mixed models

AR Gilmour; R Thompson; BR Cullis

Average information REML: An efficient algorithm for variance parameter estimation in linear mixed models

AR Gilmour^*, R Thompson, BR Cullis

^*Corresponding author for this work

Royal (Dick) School of Veterinary Studies

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

Abstract

A strategy of using an average information matrix is shown to be computationally convenient and efficient for estimating variance components by restricted maximum likelihood (REML) in the mixed linear model. Three applications are described. The motivation for the algorithm was the estimation of variance components in the analysis of wheat variety means from 1,071 experiments representing 10 years and 60 locations in New South Wales. We also apply the algorithm to the analysis of designed experiments by incomplete block analysis and spatial analysis of field experiments.

Original language	English
Pages (from-to)	1440-1450
Number of pages	11
Journal	Biometrics
Volume	51
Issue number	4
Publication status	Published - Dec 1995

Keywords / Materials (for Non-textual outputs)

EM algorithm
spatial analysis
variance components
REML
RESTRICTED MAXIMUM-LIKELIHOOD
GENERATION VARIETY TRIALS
SPARSE-MATRIX INVERSION
FIELD EXPERIMENTS
COMPONENTS

Cite this

@article{d489eab392de4c44a99b0042ed11e49c,

title = "Average information REML: An efficient algorithm for variance parameter estimation in linear mixed models",

abstract = "A strategy of using an average information matrix is shown to be computationally convenient and efficient for estimating variance components by restricted maximum likelihood (REML) in the mixed linear model. Three applications are described. The motivation for the algorithm was the estimation of variance components in the analysis of wheat variety means from 1,071 experiments representing 10 years and 60 locations in New South Wales. We also apply the algorithm to the analysis of designed experiments by incomplete block analysis and spatial analysis of field experiments.",

keywords = "EM algorithm, spatial analysis, variance components, REML, RESTRICTED MAXIMUM-LIKELIHOOD, GENERATION VARIETY TRIALS, SPARSE-MATRIX INVERSION, FIELD EXPERIMENTS, COMPONENTS",

author = "AR Gilmour and R Thompson and BR Cullis",

year = "1995",

month = dec,

language = "English",

volume = "51",

pages = "1440--1450",

journal = "Biometrics",

issn = "0006-341X",

publisher = "Oxford University Press",

number = "4",

}

TY - JOUR

T1 - Average information REML: An efficient algorithm for variance parameter estimation in linear mixed models

AU - Gilmour, AR

AU - Thompson, R

AU - Cullis, BR

PY - 1995/12

Y1 - 1995/12

N2 - A strategy of using an average information matrix is shown to be computationally convenient and efficient for estimating variance components by restricted maximum likelihood (REML) in the mixed linear model. Three applications are described. The motivation for the algorithm was the estimation of variance components in the analysis of wheat variety means from 1,071 experiments representing 10 years and 60 locations in New South Wales. We also apply the algorithm to the analysis of designed experiments by incomplete block analysis and spatial analysis of field experiments.

AB - A strategy of using an average information matrix is shown to be computationally convenient and efficient for estimating variance components by restricted maximum likelihood (REML) in the mixed linear model. Three applications are described. The motivation for the algorithm was the estimation of variance components in the analysis of wheat variety means from 1,071 experiments representing 10 years and 60 locations in New South Wales. We also apply the algorithm to the analysis of designed experiments by incomplete block analysis and spatial analysis of field experiments.

KW - EM algorithm

KW - spatial analysis

KW - variance components

KW - REML

KW - RESTRICTED MAXIMUM-LIKELIHOOD

KW - GENERATION VARIETY TRIALS

KW - SPARSE-MATRIX INVERSION

KW - FIELD EXPERIMENTS

KW - COMPONENTS

M3 - Article

SN - 0006-341X

VL - 51

SP - 1440

EP - 1450

JO - Biometrics

JF - Biometrics

IS - 4

ER -

Average information REML: An efficient algorithm for variance parameter estimation in linear mixed models

Abstract

Keywords / Materials (for Non-textual outputs)

Fingerprint

Cite this