Model Selection in Linear Mixed Models
By Samuel Müller, Janice L. Scealy, Alan H. Welsh. 2013
298 citations recorded by the source corpus.
Linear mixed effects models are highly flexible in handling a broad range of data types and are therefore widely used in applications. A key part in the analysis of data is model selection, which often aims to choose a parsimonious model with other desirable properties from a possibly very large set of candidate statistical models. Over the last 5-10 years the literature on model selection in linear mixed models has grown extremely rapidly. The problem is much more complicated than in linear regression because selection on the covariance structure is not straightforward due to computational issues and boundary problems arising from positive semidefinite constraints on covariance matrices. To obtain a better understanding of the available methods, their properties and the relationships between them, we review a large body of literature on linear mixed model selection. We arrange, implement, discuss and compare model selection methods based on four major approaches: information criteria such as AIC or BIC, shrinkage methods based on penalized loss functions such as LASSO, the Fence procedure and Bayesian techniques.
Subfields: Environmental Engineering, Statistics and Probability
View Paper: https://www.semanticscholar.org/paper/8d9f5469762406d05f210af197c4540ffa5fed6d
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 Statistics and Probability
- Arunas P. Verbyla
- William F. Fagan
- D. Doak
- K. Hodges
- Ç. Şekercioğlu
- D. Keinath
- M. Kauffman
- Laura R. Prugh
Citations in the corpus
- A note on model selection using information criteria for general linear models estimated using REML (2019, methodology): “6 of that paper) and Müller, Scealy & Welsh (2013) who suggest the REML estimates can be used with the full likelihood for the AIC and BIC (see their equation (10) and the last sentence in Section 3.”
- A global analysis of traits predicting species sensitivity to habitat fragmentation (2017, methodology): “We guarded against over-fitting by limiting model complexity to 12 terms at each step and comparing competing models using the Bayes information criterion (BIC; Burnham & Anderson, 2004; Muller et al., 2013; Aho et al., 2014).”