Regression Analysis for Multivariate Random Functions

• Activity Number: 286
• Type: Invited
• Date/Time: Tuesday, August 11, 2015 : 8:30 AM to 10:20 AM
• Sponsor: IMS
• Author(s): Jeng-Min Chiou*
• Companies: Academia Sinica
• Keywords: Functional prediction ; Functional principal component analysis ; Functional regression ; Multivariate functional data ; Transformation

Abstract

We consider a regression model in which both the response and the predictor variables contain multivariate random functions. We propose a multivariate functional linear regression model for the analysis and prediction of multivariate functional data. This model makes use of the transformed multivariate functional principal component analysis to take the advantage of functional covariance structures and accommodate incomparable magnitudes of variation between random functions within the multivariate response and predictor variables, respectively. We provide the asymptotic results for statistical inference and present numerical results, including the simulation and the data application, to demonstrate the usefulness of this method.