CMStatistics 2020: Start Registration
View Submission - CMStatistics
Title: Permutation and Bayesian tests for random effects in mixed models Authors:  Reza Drikvandi - Durham University (United Kingdom) [presenting]
Abstract: Random effects are used in mixed models to account for the unknown between-subject variability as well as the within-subject correlation in longitudinal, multilevel, clustered and other correlated data. Since random effects are latent and unobservable variables, it is challenging to decide which random effects to include or exclude from the model. In statistical language, this would be equivalent to testing whether or not the variance components of random effects are zero. However, test for zero variance components is a nonstandard testing problem because the null hypothesis in on the boundary of the parameter space and consequently the standard tests such as likelihood ratio, Wald and score tests may not be easily used. We introduce permutation tests and Bayesian tests for testing random effects which avoid the issues with the boundary of parameter space. The proposed methods are illustrated via simulations and a real data application.