B0944
Title: The joint models for longitudinal and survival data: Analysis and extensions
Authors: Marcella Mazzoleni - University of Bergamo (Italy) [presenting]
Abstract: The joint models for survival and longitudinal data became an appealing topic these years. In fact, several researchers decided to focus and to propose possible extensions of these models. The original idea of joint models was to jointly analyse the two sub-models, namely longitudinal and survival, quantifying the effect of one longitudinal covariate on the risk of an event. Starting from this point different extensions of the sub-models, several estimation methods, and various applications, most of which in the medical field, were proposed. The focus is on extending the longitudinal sub-model, analysing more than just one longitudinal covariate, appropriately adjusting the estimation method based on maximising the likelihood function through the implementation of an Expectation-Maximisation algorithm. Accordingly, the diagnostic and goodness of fit elements are updated considering more than just one longitudinal covariate, implementing the estimated survival function, and the residuals and dynamic predictions for both survival and longitudinal sub-models.
