Likelihood ratio tests in linear mixed models with one variance component

By Ciprian M. Crainiceanu, David Ruppert. 2003

483 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/383d2730ef1eaabf601a682ce52c62985d8404c7

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

Size and power of tests for a zero random effect variance or polynomial regression in additive and linear mixed models (2008, methodology): “Violation of the assumption results in a conservative test (Crainiceanu and Ruppert, 2004; Pinheiro and Bates, 2000).”
The Hodrick-Prescott Filter A Special Case of Penalized Spline Smoothing (2010, methodology): “Crainiceanu and Ruppert (2004), and Crainiceanu, Ruppert, Claeskens, and Wand (2005), develop both exact and asymptotic likelihood ratio tests for ML and REML-based inference.”
Bootstrap tests for variance components in generalized linear mixed models (2009, background): “intercept and a random cluster efiect, Crainiceanu and Ruppert (2004) showed that for”
Bayesian generalized least squares regression with application to log Pearson type 3 regional skew estimation (2005, unknown): “[3] Cunnane [1988] and Groupe de recherche en hydrologie statistique (GREHYS) [1996a, 1996b] review methods that use regional information in the estimation of hydrologic statistics.”
Likelihood Ratio Tests for Dependent Data with Applications to Longitudinal and Functional Data Analysis (2014, methodology): “For example, Fan et al. (2001) introduced a generalized likelihood ratio testing (LRT), while Crainiceanu & Ruppert (2004) and Crainiceanu et al. (2005) introduced an LRT.”
Exact likelihood ratio tests for penalised splines (2005, methodology): “For nested linear models, this difference is d (I−S 0 )Y d2−d (I−S A )Y d2, but also equals d (S A −S0 )Y d2 and dSA (I−S0 )Y d2, where S0 and SA are the hat matrices for the null and alternative models.”
Bivariate Binomial Spatial Modeling of Loa loa Prevalence in Tropical Africa (2008, methodology): “Instead we used the finite-sample distribution of the RLRT derived by Crainiceanu and Ruppert (2004) and Crainiceanu, Ruppert, Claeskens, and”
Non‐parametric small area estimation using penalized spline regression (2008, methodology): “These asymptotic approximations are very easy to compute but, as discussed in Crainiceanu and Ruppert (2004) in the context of testing for the presence of spline components, they do not perform satisfactorily in practice.”
Restricted Likelihood Ratio Testing for Zero Variance Components in Linear Mixed Models (2008, methodology): “An important conclusion is that the 0.5χ20 : 0.5χ 2 1 approximation is very conservative in all cases, as expected from Crainiceanu and Ruppert (2004b).”
Best-Practice Recommendations for Estimating Cross-Level Interaction Effects Using Multilevel Modeling (2013, background): “025, and results based on the bootstrap confidence interval, the −2 log-likelihood, and the Crainiceanu and Ruppert (2004) test suggest that this value is unlikely to be zero in the population.”
Improved detection of changes in species richness in high diversity microbial communities (2016, background): “ only know the standard error in the estimates ^˙ i and not the true standard deviations ˙ i. For this reason we base our choice of asymptotics on the results of Demidenko (2004) rather than those of Crainiceanu and Ruppert (2004). Thus our restricted ML procedure maximizes 7 l R( 2 0;;˙ u ) = 2 1 2”
Determination of linear components in additive models (2011, methodology): “The LRT and restricted likelihood ratio test (RLRT) have been investigated in Crainiceanu and Ruppert (2003) , Crainiceanu, Ruppert, and Vogelsang (2003) and Claeskens (2004) for comparing linear models and linear mixed models with one/two variance components.”
Spatially Adaptive Bayesian Penalized Splines With Heteroscedastic Errors (2007, background): “For example, Crainiceanu and Ruppert (2004) showed that, in finite samples, the (RE)ML estimator of the smoothing parameter is biased towards oversmoothing and Kauermann (2002) obtained corresponding asymptotic results for smoothing splines.”
Efficient Control of Population Structure in Model Organism Association Mapping (2008, methodology): “Our method is related to a similar technique developed in a different context of simulating the null distribution of variance-component test statistics (C rainiceanu and R uppert 2004).”