Feature Sensitive Curve Registration by Kernel Matching
Abstract
In this paper, we argue that the problem of registering two sets of functional data, where the underlying mean function has sharp features, is not properly addressed by methods designed to align a bunch of growth curves data. We provide a new method, which is able to pool local information without smoothing and to match sharp landmarks without manual identification. This method, which we refer to as kernel-matched registration, is based on maximizing a kernel-based measure of alignment. We prove that the proposed method is consistent under fairly general conditions. Simulation results show superiority of the performance of the proposed method over two existing methods. The proposed method is illustrated through the analysis of three sets of paleoclimatic data.
Keywords
Cite
@article{arxiv.1704.03127,
title = {Feature Sensitive Curve Registration by Kernel Matching},
author = {Dibyendu Bhaumik and Radhendushka Srivastava and Debasis Sengupta},
journal= {arXiv preprint arXiv:1704.03127},
year = {2017}
}
Comments
29 pages, 10 figures; Changes over the earlier version: Some of the sections are modified; a new section on "Standard error of the estimator" has been added. More analysis are reported in Simulation of performance