Various methods have been proposed for estimating the parameters of both linear and nonlinear models for repeated measurements. In this talk, we attempt to link these different methods using the principle of least squares (LS) and generalized estimating equations (GEE) as our guide. We then consider how these methods can be applied to a class of generalized nonlinear mixed-effects models including linear and generalized linear mixed-effects models. Examples from pharmacokinetic, clinical, and nonlinear growth curve applications are given which compare and contrast these different models and estimators. The relative merits of the different estimators are discussed and some general recommendations given.