We present some asymptotic results on the distance between the means of samples of curves generated by independent continuous time stochastic processes in L2(T). The asymptotic results are based on mild assumptions on the moments of the processes, and there are no conditions on their probability distribution. The metrics we consider extends the Mahalanobis distance to L2(T) without any truncation on the first principal components. Applications in the context of classification of functional data are finally discussed.

Ghiglietti, A., Ieva, F., Maria Paganoni, A., A generalized Mahalanobis distance for the classification of functional data, in Classification and Data Analysis Group: Book of abstracts, (Milano, 13-15 September 2017), Universitas Studiorum S.r.l. Casa Editrice, Mantova 2017: 1-6 [http://hdl.handle.net/10807/117863]

A generalized Mahalanobis distance for the classification of functional data

Ghiglietti, Andrea
;
2017

Abstract

We present some asymptotic results on the distance between the means of samples of curves generated by independent continuous time stochastic processes in L2(T). The asymptotic results are based on mild assumptions on the moments of the processes, and there are no conditions on their probability distribution. The metrics we consider extends the Mahalanobis distance to L2(T) without any truncation on the first principal components. Applications in the context of classification of functional data are finally discussed.
Inglese
Classification and Data Analysis Group: Book of abstracts
Classification and Data Analysis Group 2017
Milano
13-set-2017
15-set-2017
978-88-99459-71-0
Universitas Studiorum S.r.l. Casa Editrice
Ghiglietti, A., Ieva, F., Maria Paganoni, A., A generalized Mahalanobis distance for the classification of functional data, in Classification and Data Analysis Group: Book of abstracts, (Milano, 13-15 September 2017), Universitas Studiorum S.r.l. Casa Editrice, Mantova 2017: 1-6 [http://hdl.handle.net/10807/117863]
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10807/117863
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact