Nonparametric Functional Concurrent Regression Models

• Activity Number: 162
• Type: Topic Contributed
• Date/Time: Monday, August 4, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Arnab Maity*+ and Janet Kim and Ana-Maria Staicu
• Companies: North Carolina State University and North Carolina State University and North Carolina State University
• Keywords: Concurrent models ; Funtional data ; Spline modeling ; Functional principal components analysis

Abstract

We propose a generalization of the functional concurrent regression model, where both the response and the covariate are functional data, observed on the same time domain. In contrast to typical functional concurrent models, we allow the relationship between the response and covariate to be nonlinear and vary with both the current time point and the current covariate. In this framework we develop methodology for estimation and testing of the unknown relationship, where we allow for correlated error structure as well as sparse and/or irregular sampling design. The proposed method is illustrated through simulation study and application to real world data.