A scalable and efficient covariate selection criterion for mixed effects regression models with unknown random effects structure

DOI10.1016/J.CSDA.2017.07.011zbMATH Open1469.62047arXiv1612.00759OpenAlexW2995443996MaRDI QIDQ1662050FDOQ1662050

Authors: Radu V. Craiu, Thierry Duchesne 

Publication date: 17 August 2018

Published in: Computational Statistics and Data Analysis (Search for Journal in Brave)

Abstract: We propose a new model selection criterion for mixed effects regression models that is computable when the model is fitted with a two-step method, even when the structure and the distribution of the random effects are unknown. The criterion is especially useful in the early stage of the model building process when one needs to decide which covariates should be included in a mixed effects regression model but has no knowledge of the random effect structure. This is particularly relevant in substantive fields where variable selection is guided by information criteria rather than regularization. The calculation of the criterion requires only the evaluation of cluster-level log-likelihoods and does not rely on heavy numerical integration. We provide theoretical and numerical arguments to justify the method and we illustrate its usefulness by analyzing data on a socio-economic study of young American Indians.

Full work available at URL: https://arxiv.org/abs/1612.00759

Recommendations

• Selection strategy for covariance structure of random effects in linear mixed-effects models
• Generalized degrees of freedom and adaptive model selection in linear mixed-effects models
• Model selection in linear mixed effect models
• Moment-based method for random effects selection in linear mixed models
• New variable selection for linear mixed-effects models

zbMATH Keywords

variable selectionAkaike information criteriongeneralized linear mixed modeltwo-stage estimationrandom coefficient modelh-likelihood

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Generalized linear models (logistic models) (62J12)

Cites Work

• Fence methods for mixed model selection
• Estimating the dimension of a model
• Regression and time series model selection in small samples
• Asymptotic properties of criteria for selection of variables in multiple regression
• Linear and generalized linear mixed models and their applications.
• On the behaviour of marginal and conditional AIC in linear mixed models
• Fixed and Random Effects Selection in Mixed Effects Models
• Model Selection and Multimodel Inference
• Conditional Akaike information for mixed-effects models
• Pseudo-likelihood methodology for partitioned large and complex samples
• Linear mixed models for longitudinal data
• Variable selection in linear mixed effects models
• Model selection in linear mixed models
• A note on conditional AIC for linear mixed-effects models
• A pairwise likelihood approach to estimation in multilevel probit models
• Title not available (Why is that?)
• Statistical predictor identification
• Unifying the derivations for the Akaike and corrected Akaike information criteria.
• Conditional Akaike information under generalized linear and proportional hazards mixed models
• Information based model selection criteria for generalized linear mixed models with unknown variance component parameters
• A strongly consistent procedure for model selection in a regression problem
• A note on BIC in mixed-effects models
• A unifying approach to the estimation of the conditional Akaike information in generalized linear mixed models
• A note on conditional Akaike information for Poisson regression with random effects
• Title not available (Why is that?)

Cited In (2)

• Focused model selection for linear mixed models with an application to whale ecology
• Consistent Fixed-Effects Selection in Ultra-high dimensional Linear Mixed Models with Error-Covariate Endogeneity

Uses Software

• lme4
• TMB
• cAIC4

This page was built for publication: A scalable and efficient covariate selection criterion for mixed effects regression models with unknown random effects structure

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1662050)