Detecting Influential Measurements in a Growth Curves Model

This paper comprises a study of the influence of deleting measurements on growth curve estimates. The problem of influential measurements is considered both at the inference level and at the design level. A measure of detecting influential vs uninfluential measurements is proposed. The influence analysis envisaged also serves as a means for comparing the robustness of various models to missing measurements and to different study designs. The paper concludes with an illustrative application.