Model comparison with composite likelihood information criteria

DOI10.3150/13-BEJ539zbMATH Open1357.62086arXiv1410.4354OpenAlexW3098209628MaRDI QIDQ470048FDOQ470048

Authors: Chi Tim Ng, Harry Joe 

Publication date: 11 November 2014

Published in: Bernoulli (Search for Journal in Brave)

Abstract: Comparisons are made for the amount of agreement of the composite likelihood information criteria and their full likelihood counterparts when making decisions among the fits of different models, and some properties of penalty term for composite likelihood information criteria are obtained. Asymptotic theory is given for the case when a simpler model is nested within a bigger model, and the bigger model approaches the simpler model under a sequence of local alternatives. Composite likelihood can more or less frequently choose the bigger model, depending on the direction of local alternatives; in the former case, composite likelihood has more "power" to choose the bigger model. The behaviors of the information criteria are illustrated via theory and simulation examples of the Gaussian linear mixed-effects model.

Full work available at URL: https://arxiv.org/abs/1410.4354

Recommendations

• Composite likelihood Bayesian information criteria for model selection in high-dimensional data
• A note on composite likelihood inference and model selection
• Evaluating the relative merits of competing models based on empirical likelihood ratio test
• A COMPARATIVE STUDY OF INFORMATION CRITERIA FOR MODEL SELECTION
• scientific article; zbMATH DE number 4100431
• Model Selection and Information Criterion
• Publication:4939083
• Issues and strategies in the selection of composite likelihoods
• An information criterion for likelihood selection
• A modified information criterion for model selection

zbMATH Keywords

local alternativesmodel comparisonAkaike information criterionBayesian information criterionmixed-effects model

Mathematics Subject Classification ID

Parametric hypothesis testing (62F03) Asymptotic properties of parametric estimators (62F12) Bayesian inference (62F15)

Cites Work

• Mixed-Effects Models in S and S-PLUS
• On weighting of bivariate margins in pairwise likelihood
• Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses
• Random-Effects Models for Longitudinal Data
• Title not available (Why is that?)
• A note on composite likelihood inference and model selection
• Title not available (Why is that?)
• Title not available (Why is that?)
• On composite marginal likelihoods
• Distribution of Quadratic Forms in Normal Random Variables—Evaluation by Numerical Integration
• Semiparametric Models for Longitudinal Data with Application to CD4 Cell Numbers in HIV Seroconverters
• Title not available (Why is that?)
• A general family of limited information goodness-of-fit statistics for multinomial data
• An Elementary Proof of Binet's Formula for the Gamma Function
• Composite likelihood Bayesian information criteria for model selection in high-dimensional data
• On the robustness of maximum composite likelihood estimate

Cited In (5)

• Non-stationary dependence structures for spatial extremes
• Model selection by pathwise marginal likelihood thresholding
• A note on composite likelihood inference and model selection
• Inference and forecasting for continuous-time integer-valued trawl processes
• On the theory of periodic multivariate INAR processes

Uses Software

• S-PLUS
• MEMSS

This page was built for publication: Model comparison with composite likelihood information criteria

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q470048)