Title: Bayesian two-part hidden Markov models for semicontinuous longitudinal data with nonignorable missing covariates
Authors: Xiaoxiao Zhou - The Chinese University of HongKong (Hong Kong) [presenting]
Abstract: A two-part hidden Markov Model is developed for the analysis of semicontinuous longitudinal data. The two-part model manages a semicontinuous variable by splitting it into two random variables, a binary indicator to determine the occurrence of excess zeros at all occasions, and a continuous random variable to determine its actual level. For the continuous longitudinal response, a hidden Markov model (HMM) is proposed further to describe the relationship between the observation process and the unobservable finite-state transition process. The HMM consists of two major components. The first component is a transition model for investigating how potential covariates influence the probabilities of transitioning from one hidden state to another. The second component is a conditional regression model for examining the effects of covariates on the response. A full Bayesian approach together with efficient Markov chain Monte Carlo methods is developed for statistical inference in the presence of missing covariates. The proposed methodology is applied to a study on the Alzheimers Disease Neuroimaging data set. New sights into the pathology of Alzheimers disease and its potential risk factors are obtained.