Bayesian Functional Regression Model for Analyzing Intracranial Pressure Data

• Activity Number: 140
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Lu Wang*+ and Donatello Telesca
• Companies: University of California, Los Angeles and University of California at Los Angeles
• Keywords: functional regression ; MCMC ; repeated measurement ; curve registration

Abstract

Functional data can be measured repeatedly, which brings the challenge to analyze longitudinal functional data. One such motivating study comes from intensive-care unit (ICU), where intracranial pressure (ICP) is monitored for patients who have severe brain damage. We build this functional regression model under bayesian framework to investigate the relationship between ICP and clinical outcome of these patients. Considering the longitudinal nature of the data, we integrate curve registration step into our joint model to account for the between-subject and within-subject variability. Given the relative high dimension of parameters (each patient has average 200 repeated measurements), we also adopt advanced computation method for MCMC simulation.