CMStatistics 2017: Start Registration
View Submission - CMStatistics
Title: Mixed effects modeling and warping for functional data using B-spline Authors:  Emilie Devijver - CNRS (France) [presenting]
Gerda Claeskens - KU Leuven (Belgium)
Irene Gijbels - Katholieke Universiteit Leuven (Belgium)
Abstract: In functional data the interest is to find a global mean pattern, but also to capture the individual curve differences in phase and amplitude. This can be done conveniently by building in random effects on two levels: a random effect appearing in the warping functions for accounting individual phase variations; and a random effect to deal with individual amplitude variations. Via an appropriate choice of the warping function and B-spline approximations, estimation in the non-linear mixed effects functional model is feasible, and does not require any prior knowledge on landmarks for the functional data. Sufficient and necessary conditions for identifiability of the flexible model are provided. The finite-sample performance of the proposed estimation procedure is investigated in a simulation study, which includes comparisons with an existing method. The added value of the developed method is further illustrated on the analysis of a real data example.