TESTING FOR HETEROGENEITY IN NONLINEAR MODELS WITH RANDOM EFFECTS
非线性随机效应模型的异方差性检验

Abstract Random-effects models are widely used to characterize repeated measurementdata. Random-effect models for longitudinal data include both within-and betweensubjectvariances. Zhang and Weiss (2000) studied heterogeneity tests for linear models with randomeffects. In this paper, three types of heterogeneity of variance are discussed in nonlinear modelswith random effects. They are variance heterogeneity of within, between and within-betweenmultivariate subjects. Several score test statistics are obtained and the relations of three scorefunctions are presented. A numerical example is given to illustnde the methodology