RT Generic
T1 Spatial depth-based classification for functional data
A1 Sguera, Carlo
A1 Galeano San Miguel, Pedro
A1 Lillo Rodríguez, Rosa Elvira
A2 Universidad Carlos III de Madrid. Departamento de Estadística, 
AB Functional data are becoming increasingly available and tractable because of the lasttechnological advances. We enlarge the number of functional depths by defining twonew depth functions for curves. Both depths are based on a spatial approach: thefunctional spatial depth (FSD), that shows an interesting connection with the functionalextension of the notion of spatial quantiles, and the kernelized functional spatial depth(KFSD), which is useful for studying functional samples that require an analysis at alocal level. Afterwards, we consider supervised functional classification problems, andin particular we focus on cases in which the samples may contain outlying curves. Forthese situations, some robust methods based on the use of functional depths areavailable. By means of a simulation study, we show how FSD and KFSD perform asdepth functions for these depth-based methods. The results indicate that a spatial depthbasedclassification approach may result helpful when the datasets are contaminated,and that in general it is stable and satisfactory if compared with a benchmark proceduresuch as the functional k-nearest neighbor classifier. Finally, we also illustrate ourapproach with a real dataset.
YR 2012
FD 2012-05
LK https://hdl.handle.net/10016/14331
UL https://hdl.handle.net/10016/14331
LA eng
NO This research was partially supported bySpanish Ministry of Education and Science grant 2007/04438/001, by Madrid Regiongrant 2011/00068/001, by Spanish Ministry of Science and Innovation grant2012/00084/001 and by MCI grant MTM2008-03010.
DS e-Archivo
RD 19 may. 2024