Fence methods for mixed model selection
From MaRDI portal
Abstract: Many model search strategies involve trading off model fit with model complexity in a penalized goodness of fit measure. Asymptotic properties for these types of procedures in settings like linear regression and ARMA time series have been studied, but these do not naturally extend to nonstandard situations such as mixed effects models, where simple definition of the sample size is not meaningful. This paper introduces a new class of strategies, known as fence methods, for mixed model selection, which includes linear and generalized linear mixed models. The idea involves a procedure to isolate a subgroup of what are known as correct models (of which the optimal model is a member). This is accomplished by constructing a statistical fence, or barrier, to carefully eliminate incorrect models. Once the fence is constructed, the optimal model is selected from among those within the fence according to a criterion which can be made flexible. In addition, we propose two variations of the fence. The first is a stepwise procedure to handle situations of many predictors; the second is an adaptive approach for choosing a tuning constant. We give sufficient conditions for consistency of fence and its variations, a desirable property for a good model selection procedure. The methods are illustrated through simulation studies and real data analysis.
Recommendations
Cites work
- scientific article; zbMATH DE number 47282 (Why is no real title available?)
- scientific article; zbMATH DE number 3635352 (Why is no real title available?)
- scientific article; zbMATH DE number 2017771 (Why is no real title available?)
- scientific article; zbMATH DE number 3444596 (Why is no real title available?)
- scientific article; zbMATH DE number 1907572 (Why is no real title available?)
- A new look at the statistical model identification
- Approximate efficiency of a selection procedure for the number of regression variables
- Asymptotic properties of criteria for selection of variables in multiple regression
- Asymptotic properties of maximum likelihood estimates in the mixed model of the analysis of variance
- Conditional Akaike information for mixed-effects models
- Consistent procedures for mixed linear model selection
- Counting degrees of freedom in hierarchical and other richly-parameterised models
- Estimating the dimension of a model
- Fence methods for mixed model selection
- Improved Estimation Procedures for Multilevel Models with Binary Response: A Case-Study
- Maximum Likelihood Approaches to Variance Component Estimation and to Related Problems
- Maximum-likelihood estimation for the mixed analysis of variance model
- On Measuring and Correcting the Effects of Data Mining and Model Selection
- Robust estimation in generalized linear mixed models
- The pigeonhole bootstrap
Cited in
(46)- Confidence graphs for graphical model selection
- A discussion of prior-based Bayesian information criterion (PBIC)
- A note on BIC in mixed-effects models
- The fence methods
- A Relaxation Approach to Feature Selection for Linear Mixed Effects Models
- A scalable and efficient covariate selection criterion for mixed effects regression models with unknown random effects structure
- Small area estimation via heteroscedastic nested-error regression
- A connection between the observed best prediction and the fence model selection method
- Globaltest confidence regions and their application to ridge regression
- FBST for Mixture Model Selection
- Observed best selective prediction in small area estimation
- Restricted fence method for covariate selection in longitudinal data analysis
- Model selection in linear mixed-effect models
- Information criteria for Fay-Herriot model selection
- Forward stability and model path selection
- Joint variable selection for fixed and random effects in linear mixed-effects models
- Variable selection for linear mixed models with applications in small area estimation
- Selection of fixed effects in high-dimensional generalized linear mixed models
- Choice of generalized linear mixed models using predictive crossvalidation
- Cross-validation with confidence
- A simplified adaptive fence procedure
- Fence methods for mixed model selection
- Robust variable selection in linear mixed models
- Subset Selection for Linear Mixed Models
- mplot
- Robust model selection with covariables missing at random
- Rejoinder
- Conditional Akaike information criterion in the Fay-Herriot model
- Model selection in linear mixed models
- Covariate selection for multilevel models with missing data
- Comments on: Goodness-of-fit tests in mixed models
- New important developments in small area estimation
- New variable selection for linear mixed-effects models
- Invisible fence methods and the identification of differentially expressed gene sets
- Joint modeling of playing time and purchase propensity in massively multiplayer online role-playing games using crossed random effects
- Small area estimation with mixed models: a review
- Asymptotic theory of generalized information criterion for geostatistical regression model selection
- Selection of linear mixed‐effects models for clustered data
- Focused model selection for linear mixed models with an application to whale ecology
- Earthquake Risk Embedded in Property Prices: Evidence From Five Japanese Cities
- Data mining for longitudinal data under multicollinearity and time dependence using penalized generalized estimating equations
- Fence methods for backcross experiments
- fence
- Robust variable selection in semiparametric mixed effects longitudinal data models
- A longitudinal study of children's aggressive behaviours based on multivariate mixed models with incomplete data
- A simple adaptation of variable selection software for regression models to select variables in nested error regression models
This page was built for publication: Fence methods for mixed model selection
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q124080)
