Litcius/Paper detail

Asymptotic Equivalence between Cross-Validations and Akaike Information Criteria in Mixed-Effects Models

Yixin Fang

2021Journal of Data Science145 citationsDOIOpen Access PDF

Abstract

For model selection in mixed effects models, Vaida and Blan chard (2005) demonstrated that the marginal Akaike information criterion is appropriate as to the questions regarding the population and the conditional Akaike information criterion is appropriate as to the questions regarding the particular clusters in the data. This article shows that the marginal Akaike information criterion is asymptotically equivalent to the leave-one-cluster-out cross-validation and the conditional Akaike information criterion is asymptotically equivalent to the leave-one-observation-out cross-validation.

Topics & Concepts

Akaike information criterionBayesian information criterionMathematicsEquivalence (formal languages)StatisticsModel selectionPopulationSelection (genetic algorithm)Information CriteriaEconometricsApplied mathematicsComputer scienceDiscrete mathematicsArtificial intelligenceSociologyDemographyStatistical Methods and Bayesian InferenceStatistical Methods and InferenceBayesian Methods and Mixture Models
Asymptotic Equivalence between Cross-Validations and Akaike Information Criteria in Mixed-Effects Models | Litcius