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

By Yixin Fang. 2021

68 citations recorded by the source corpus.

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 con- ditional Akaike information criterion is appropriate as to the questions re- garding 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 informa- tion criterion is asymptotically equivalent to the leave-one-observation-out cross-validation.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/05b755c14b8878ed362f3e600467bba990742e6e