Title: A joint model for truncated and mixed types of longitudinal and survival data
Authors: Lang Wu - University of British Columbia (Canada) [presenting]
Abstract: In the analysis of longitudinal data and survival data, joint models are useful since the longitudinal data and survival data are often strongly associated. In practice, the longitudinal data can be highly complicated, such as being truncated and mixed types of discrete and continuous. We will discuss some recent work to address these data complications in joint models. Another challenge for joint models is computation, since the likelihoods of joint models often involve high-dimensional and intractable integrations. We will also discuss a computationally efficient approximate likelihood method. The models and methods will be applied to the analysis of a recent HIV vaccine dataset.