Varying-Coefficient Additive Model for Functional Data

• Activity Number: 154
• Type: Invited
• Date/Time: Monday, August 5, 2013 : 10:30 AM to 12:20 PM
• Sponsor: International Chinese Statistical Association
• Author(s): Jane-Ling Wang*+ and Xiaoke Zhang
• Companies: UC Davis and University of California Davis
• Keywords: nonparametric smoothing ; B-splines ; regression analysis ; stochastic processes ; L^2 convergence ; fMRI data

Abstract

Both Varying-coefficient and additive models have been studied extensively in the literature as extensions to linear models. They has also been extended to functional response data. However, existing extensions are still restrictive to refelct the functional feature of the responses. In this paper, we extend both models to a much more flexible one, termed "varying-coefficient additive model" and propose a simple algorithm to estimate the nonparametric varying coefficients and the nonparametric additive functions. We establish L^2 rates of convergence for each component function and demonstrate the applicability of the new model to fMRI data.