|
Abstract:
|
In the longitudinal study of neurodegenerative disorders such as Alzheimer's disease (AD) and Parkinson's disease (PD), multiple longitudinal health outcomes (continuous and categorical) are measured because these diseases are heterogeneous disorders characterized by multiple impaired domains with variable clinical symptoms and disease progression. Missing data are ubiquitous in these studies due to missed visits, withdrawals, lost to follow-up, death, etc. In this talk, we will first review the state-of-the-art methods in handling missing data in these multivariate longitudinal data structure. We will then present several methods based on latent variable modeling and functional principal component analysis and apply these methods to multiple real studies of AD and PD.
|
