1988
DOI: 10.1080/01621459.1988.10478693
|View full text |Cite
|
Sign up to set email alerts
|

Newton—Raphson and EM Algorithms for Linear Mixed-Effects Models for Repeated-Measures Data

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
232
0
4

Year Published

1996
1996
2017
2017

Publication Types

Select...
6
3

Relationship

1
8

Authors

Journals

citations
Cited by 446 publications
(240 citation statements)
references
References 8 publications
0
232
0
4
Order By: Relevance
“…Detailed discussions of these iterative computer algorithms are beyond the scope of this paper, and we refer the interested reader to Lindstrom and Bates. 28 The introduction of SAS Proc MIXED 29 has greatly facilitated the implementation of the mixed model for general practitioners.…”
Section: Definitions and Assumptionsmentioning
confidence: 99%
“…Detailed discussions of these iterative computer algorithms are beyond the scope of this paper, and we refer the interested reader to Lindstrom and Bates. 28 The introduction of SAS Proc MIXED 29 has greatly facilitated the implementation of the mixed model for general practitioners.…”
Section: Definitions and Assumptionsmentioning
confidence: 99%
“…Os pacotes R utilizados para modelagem foram "nlme" e "multilevel", especialmente a função "lme()", que se fundamenta na estrutura descrita por Lindstrom e Bates (1988). Também foram realizados testes com o pacote "lme4", mas ele se mostrou menos robusto quando se adicionou a variável de peso.…”
Section: Análise Dos Dadosunclassified
“…Setting 0 to be the upper triangular elements of L gives the Cholesky parametrization of 27. Lindstrom and Bates (1988) use this parametrization to obtain derivatives of the log-likelihood of a linear mixed-effects model for use in a Newton-Raphson algorithm. They reported that the use of this parametrization dramatically improved the convergence properties of the optimization algorithm, when compared to a constrained estimation approach.…”
Section: Cholesky Parametrizationmentioning
confidence: 99%
“…Lindstrom and Bates (1988) have shown that the log-likelihood in (5) can be profiled to produce a function of 27 alone. In the simulation, we used 27 matrices of dimensions 3 and 6.…”
Section: Fig 2 Box-plots Of User Time and Number Of Iterations To Cmentioning
confidence: 99%