Advanced statistics for longitudinal analyses

Chief Investigators

Funded by: Monash University Strategic Grants Scheme

This project aims to apply the advanced statistical analysis method, namely Growth Mixture Models, to data from epidemiological longitudinal studies. The purpose of a mixture modelling analysis is to estimate the number of underlying subpopulations and to assign to each individual a probability of belonging to each subpopulation. These advanced analysis techniques have not been widely used in practice because (1) their mathematical theory is difficult to understand and (2) their computing methods have not been implemented in the widely used statistical packages, such as SAS and Stata. Through this project, we will apply these methods to longitudinal data and document the analysis procedure in such a way as to make it available to a wider research community. This will be a very useful tool for researchers who work with longitudinal data in epidemiological or clinical research.