Event

Trajectory modelling provides a set of tools to analyse individual longitudinal data with a goal of yielding clusters of people who share 'similar' structure in the evolution of a variable of interest over time. In addition, the methods allow identification of covariates associated with separate cluster membership. These techniques depend on notions of similarity/dissimilarity between vector objects. I will briefly describe two distinct approaches to trajectory modelling, both of which are driven by the 'level' of the longitudinal variable to a great extent. I will discuss a recent alternative approach developed by my student, Brianna Heggeseth, that uses a different measure of similarity focusing on the 'shape' of the trajectories. The ideas will be motivated and applied to childhood growth patterns using data from the Centre for the Health Assessment of Mothers and Children of Salinas (CHAMACOS) study.