A0384
Title: The two sample problem for functional data
Authors: Manuel Febrero-Bande - University of Santiago de Compostela (Spain) [presenting]
Wenceslao Gonzalez-Manteiga - University of Santiago de Compostela (Spain)
Ana Colubi - University of Giessen (Germany)
Gil Gonzalez-Rodriguez - University of Oviedo (Spain)
Abstract: The aim is to present a new test for the two-sample problem, i.e. when it is reasonable to conclude that two samples (possibly with different sizes) belong to the same distribution. The two-sample problem is solved in univariate samples through the use of the distribution functions that are not readily available in functional data scenarios. The proposal is based on energy statistics and can be applied (calibrated) to all scenarios including functional data analysis. Previous approaches in this context will be compared against the new proposal through simulation studies.
