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

By Youngjo Lee, John A. Nelder. 2001

376 citations recorded by the source corpus.

Subfields: Environmental Engineering, Plant Science, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/8934068d8303d95bf661595b8f8d351d793ea6e3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Environmental Engineering

In Plant Science

In Statistics and Probability

Citations in the corpus

The R Package geepack for Generalized Estimating Equations (2005, background): “An alternative generalization was proposed by Lee and Nelder (1996, 2001).”
Small Area Estimation Under Linear and GeneralizedLinear Mixed Models With Time and Area Effects (2003, methodology): “Lee and Nelder (2001a, 2001b) further extend the work of Nelder and Lee (1996) to correlated non-normal data.”
Modeling Heterogeneous Variance–Covariance Components in Two-Level Models (2014, methodology): “To test hypotheses about the sources of Level-1 heterogeneity, the mean and variance functions should be modeled jointly (Aitkin, 1999; Lee & Nelder, 1996, 2001).”
Random Coefficient Models (1994, methodology): “The Fisher scoring algorithm for maximising the log-likelihood in (5) is described in the Appendix; for details and applications, see see Longford (1993), and for an alternative method Goldstein (2000).”
Generalized additive models for location, scale and shape (2005, methodology): “Lee and Nelder (1996, 2001a) assumed independent random effects, whereas Lee and Nelder (2001b) relaxed this assumption to allow correlated random effects.”
A Family of Generalized Linear Models for Repeated Measures With Normal and Conjugate Random Effects (pp.188) (2010, background): “Important unifying and computational progress has been made by Lee and Nelder (1996, 2001a, 2001b, 2003) (see also Lee, Nelder and Pawitan, 2006) by proposing so-called hierarchical generalized linear models, offering a broad class of outcome and random-effects distributions, combined with…”
A pairwise likelihood approach to generalized linear models with crossed random effects (2005, methodology): “We adopt the same parameterization of fixed effects used by Lin and Breslow (1996). Table 1 displays the mean values of the parameter estimates (simulated standard errors in brackets) over 1000 trials.”
Double hierarchical generalized linear models (2006, methodology): “This method works generally well in various models: see the simulation studies of Poisson and binomial models with moderately large binomial denominator (Lee and Nelder, 2001a), of frailty models (Ha et al., 2001) and of mixed linear models with censoring (Ha et al., 2002).”
Bayesian measures of model complexity and fit (2002, methodology): “X;Z/V{.α;β/|y}] ≈ tr.VÅV−1/; where VÅ = ( XTW−1X XTW−1Z ZTW−1X ZTW−1Z ) ; V = ( XTW−1X XTW−1Z ZTW−1X ZTW−1Z +D−1 ) : This matches the proposal of Lee and Nelder (1996) except their D−1 is a diagonal matrix of the second derivatives of the prior likelihood for each random effect.”
Generalized beta regression models for random loss given default (2008, methodology): “Alternatively, the marginal likelihood can be approximated analytically by the use of the Laplace approximation to the integral, such as the penalized quasi-likelihood (QL) estimation (Breslow and Clayton (1993)) and the h-likelihood (Lee and Nelder (2001)), thereby avoiding numerical integration.”
When Employees Are Out of Step with Coworkers: How Job Satisfaction Trajectory and Dispersion Influence Individual- and Unit-Level Voluntary Turnover (2012, methodology): “The Bernoulli model in HGLM, which uses a binomial sampling method with a Bernoulli distribution and logit link (Raudenbush et al., 2004), is thus ideal for testing multilevel models with binary outcome variables (Guo & Zhao, 2000; Lee & Nelder, 2001).”
Double hierarchical generalized linear models (with discussion) (2006, methodology): “This method works generally well in various models: see the simulation studies of Poisson and binomial models with moderately large binomial denominator (Lee and Nelder, 2001a), of frailty models (Ha et al., 2001) and of mixed linear models with censoring (Ha et al., 2002).”
MODELLING CATCH AND EFFORT DATA USING GENERALISED LINEAR MODELS, THE TWEEDIE DISTRIBUTION, RANDOM VESSEL EFFECTS AND RANDOM STRATUM-BY-YEAR EFFECTS (2004, neutral): “Punt et al. (2000) and Brandão et al.”