B1639
Title: Flexible models for longitudinal data
Authors: Helen Ogden - University of Southampton (United Kingdom) [presenting]
Abstract: Models for longitudinal data are discussed, where the data consists of noisy measurements taken at several different time points for each individual, and the aim is to model how each individual's underlying response varies over time. If linear variation of the responses over time is assumed, a linear mixed model can be used for this task. More flexible modelling approaches are discussed, which allow the variation of the response over time to be any smooth curve. There is a strong link between models for functional data. Previous work on adapting methods designed for functional data is described (where measurements are typically taken very frequently) to longitudinal data (with typically only a few measurements on each individual). Some shortcomings of existing approaches are described in some examples, as well as a new methodology to solve these problems.
