Academic
Publications
Hierarchical generalised linear models: A synthesis of generalised linear models, random-effect models and structured dispersions

Hierarchical generalised linear models: A synthesis of generalised linear models, random-effect models and structured dispersions,10.1093/biomet/88.4.

Hierarchical generalised linear models: A synthesis of generalised linear models, random-effect models and structured dispersions   (Citations: 52)
BibTex | RIS | RefWorks Download
Journal: Biometrika , vol. 88, no. 4, pp. 987-1006, 2001
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...For hierarchical generalized linear models, the hierarchical likelihood (or hlikelihood), was proposed by Lee and Nelder [32]; see also [33, 34]...

    Daniel Commenges. Statistical models: Conventional, penalized and hierarchical likelihoo...

    • ...In the case where the spatial dependencies and ranges do not vary significantly but the variance does, then Hierarchical Generalized Linear Models (Lee and Nelder, 2001) might be suitable...

    R. Corstanjeet al. Inferences from fluctuations in the local variogram about the assumpti...

    • ...Our approach is somewhat different from Lee and Nelder's (Lee and Nelder 2001) hierarchical quasi-generalised linear model...

    Changchun Xieet al. Quasi-Likelihood Approach to Bioavailability and Bioequivalence Analys...

    • ...In general, let ‘ be a likelihood with nuisance effects w. Lee and Nelder (1996, 2001b) used the adjusted profile likelihood...
    • ...Note that pu(h) is the firstorder Laplace approximation to marginal likelihood m and that it would be natural to use pu(h) when m is numerically hard to obtain (Lee and Nelder 2001b)...

    Il Do Haet al. Genetic Mixed Linear Models for Twin Survival Data

    • ...Recently, normal random-effect models have been extended and for these extended models various likelihood-type estimators have been proposed and developed in statistical software; for example, the penalized quasilikelihood (PQL) procedure (GLIMMIX procedure in SAS, Breslow and Clayton 1993) for generalized linear mixed models (GLMMs), the h-likelihood procedure (HGANALYSIS procedure in GENSTAT, Lee and Nelder 1996, 2001a) for hierarchical ...
    • ...One advantage of assuming a model with independent random components is model checking because most modelchecking procedures assume independence (Lee and Nelder, 2001a)...
    • ...Lee and Nelder (1996, 2001a) considered a function pα(l), defined by...
    • ...The PQL estimators for dispersion parameters can be obtained by ignoring (possibly important) terms ∂ ˆ v/∂� in the h-likelihood REML estimation (Lee and Nelder 2001a)...

    Youngjo Leeet al. Fitting via alternative random-effect models

Sort by: