A Review on Mixed Models

[摘要] Mixed model/mixed modeling [1,2] is an important area/tool instatistics. It includes fixed effects and random effects. In fact, randomeffects (mixed) models were introduced by Fisher [3] where thecorrelations of trait values between relatives were studied. One-wayanalysis of variance (ANOVA) model and two way ANOVA are twoordinary and widely-used mixed models. Now two kinds of mixedmodels are mainly mentioned in literatures. One is to model clustereddata/repeated data/longitudinal data [4] where the response maybe divided into independent sub-vectors and the covariance matrixof random effects is very general, the other is similar to the two-wayANOVA model and some is to act as a representation tool for thenonparametric function [5,6] where the response may not be dividedinto independent sub-vectors and the covariance matrix of randomeffects is usually with special structures.