Title: Extensions on joint models for multivariate non-linear longitudinal and survival data
Authors: Ipek Guler - KU Leuven (Belgium) [presenting]
Abstract: Many follow up studies in biomedical research produce both repeated measurements and time-to-event analysis and, commonly, it is of interest to explore the association between them. For this aim, many statistical developments have been proposed to jointly model longitudinal and survival data. Further from an association between a single longitudinal biomarker and survival data, there are many extensions on multivariate longitudinal and multivariate survival data using either maximum likelihood or Bayesian approaches. However, in many situations, the computational cost is getting higher in case of having an increased number of outcomes so then the dimensional increase on the random effects covariance matrix. We will focus on the pairwise approach for multivariate longitudinal data and its use on joint modelling extensions. We will illustrate the different model approaches proposed in the literature with real biomedical data.