Conditional Akaike information for mixed-effects models

By Florin Vaida, S. Blanchard. 2005

599 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/a8df23f0769968ef0ff1ed14e1a167911839ed10

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

Model averaging and weight choice in linear mixed-effects models (2014, methodology): “Table 3 presents the values of conditional AIC, AIC and BIC for some candidate models, and weights of three model averaging methods.”
Estimating the Integrated Likelihood via Posterior Simulation Using the Harmonic Mean Identity (2006, methodology): “As described in Spiegelhalter et al. (2002), this measure has been derived from numerous perspectives, in particular in a recent extension of AIC to normal mixed models (Vaida and Blanchard, 2005).”
Can multiscale models of species’ distribution be generalized from region to region? A case study of the koala (2008, methodology): “This formulation of AIC, using the marginal likelihood, is appropriate for mixed-effects models where the modelling focus is at the population level, as in this study ( sensu Vaida & Blanchard 2005 ).”
Goodness-of-fit tests in mixed models (2009, methodology): “A value of AIC for models with random effects (with a single normal distribution) is studied by Vaida and Blanchard (2005). A ‘marginal AIC’ is there formed by considering the mixed model as a linear model though with a correlation structure for the errors that is determined by the random effects.”
Using information theory as a substitute for stepwise regression in ecology and behavior (2010, unknown): “Furthermore, there are warnings in the literature that AIC or even maximum likelihood criteria as a whole cannot be safely used in case of nested models, mixed models, and correlated parameters (Vaida and Blanchard 2005; Nakagawa et al. 2007; Ward 2008), all of which are common in our field.”
A brief introduction to mixed effects modelling and multi-model inference in ecology (2018, background): “Vaida & Blanchard (2005) provide details on AIC should be implemented for the analysis of clustered data.”
Modelling wildlife–human relationships for social species with mixed‐effects resource selection models (2007, background): “For conditional inferences within only a pack/wolf, conditional AIC, or cAIC, is recommended by Vaida & Blanchard (2005) .”
Generalized linear mixed models for phylogenetic analyses of community structure (2011, methodology): “…hypotheses about a given data set that do not involve the magnitude (or difference from zero) of a speciﬁc parameter, then model selection (e.g., with Akaike’s Information Criterion) can be used to compare different models (Akaike 1973, Burnham and Anderson 2002, Vaida and Blanchard 2005).”
Asymptotic Equivalence between Cross-Validations and Akaike Information Criteria in Mixed-Effects Models (2021, methodology): “Lemma 1 was proved in Hodges and Sargent (2001) and was used in Vaida and Blanchard (2005) and Liang et al. (2008).”
A note on BIC in mixed-effects models (2014, methodology): “In the Akaike Information criteria (AIC) adjusted to mixed models [16, 29], the log-likelihood is penalized by a term that depends on the number of degrees of freedom ρ of the model.”
Dynamic Structural Equation Models (2018, background): “…as a parameter, it is not a part of the vector θ and the likelihood used in the definition of the deviance is the marginal likelihood; that is, the latent variable has to be integrated out (for a similar discussion in the context of the Akaike’s information criterion, see Vaida & Blanchard, 2005).”
Model Selection in Linear Mixed Models (2013, methodology): “The effective degrees of freedom satisfies p≤ ρ(τ )≤ p+ s so lies between the degrees of freedom of the regression model without v and the regression model treating v as fixed effects (Vaida and Blanchard (2005)).”
Fitting Linear Mixed-Effects Models Using lme4 (2014, methodology): “(66) The trace of the hat matrix is often used as a measure of the effective degrees of freedom (e.g., Vaida and Blanchard 2005).”
A statistical framework to explore ontogenetic growth variation among individuals and populations: a marine fish example (2015, background): “…is being made at population or marginal level and each random effect can be assigned one df. Traditional, or marginal AIC, as applied here is appropriate (Vaida and Blanchard 2005), and random effects can be seen as just modeling underlying correlation structures (Claeskens and Hjort 2008:270).”