A Hierarchical Bayesian Approach to Modeling Multivariate Nonlinear Longitudinal Data: Visual World Paradigm

• Activity Number: 686
• Type: Contributed
• Date/Time: Thursday, August 13, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Melissa Pugh* and Jacob Oleson
• Companies: The University of Iowa and The University of Iowa
• Keywords: Bayesian ; Longitudinal

Abstract

Studies that focus on monitoring the change in proportion scores over time often display nonlinear temporal trends. Our work focuses on such a study where each study participant has multiple non-linear temporal trends. The hierarchical Bayesian approach allows for a convenient framework to incorporate multiple subject specific random effects, account for the correlated nature of the data, and simultaneously fit multiple correlated growth curves. The motivating dataset involving the visual world paradigm will be presented to showcase the utility of this model. The visual world paradigm is designed to assess the real-time unfolding of language processing while subjects are engaged in everyday tasks.