Asymptotics, finite-sample comparisons and applications for two-sample tests with functional data

Abstract

We consider two-sample tests for functional data with observations which may be uni- or multi-dimensional. The new methods are formulated as L-2-type criteria based on empirical characteristic functions and are convenient from the computational point of view.

Asymptotic properties are presented. Simulations and two real data applications are conducted in order to evaluate the performance of the proposed tests vis-a-vis other methods. (C) 2018 Elsevier Inc.

All rights reserved.

Keywords

• Empirical characteristic function
• Functional data
• Two-sample problem