Estimation of the law of random shifts deformation
Ismael Castillo and Jean-Michel Loubes
Preprint 2007.

Abstract

The observations are discrete values of functions which differ from an unknown feature by translation effects. These unobserved translation parameters are i.i.d. realizations of an unknown distribution, which models the variability of the response of each individual. Our aim is to construct a nonparametric estimator of the law of these random translation deformations. We provide both rates of convergence for this inverse problem and an algorithm to construct the estimator.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.

[Edit]