Date Added: Jan 2010
Longitudinal data, or data that are repeated measurements on various subjects across time, are commonplace in everyday life. Multi-level mixed models are often used for analyzing longitudinal data and drawing meaningful inferences about them. This paper discusses two common mixed models, the linear growth model and the logistic growth model, and fits them to a prototypical example that involves repeated measures on forest growth. Parameter estimates and model fitting results from two analyses are compared. The nonlinear logistic growth curve is selected as the suitable model for the current data, even though evidence from model fit statistics seems to suggest otherwise. Computer implementation is via PROC NLMIXED in the SAS 9.2 program.