The object of these short notes is to give a set of convenient symbols to define the sample space for the different compositional vectors that can be arranged into a three-way array. For the exploratory analysis of three-way data, Parafac/Candecomp and Tucker3 are some of the most applied models for low-rank decomposition of three-way arrays. Here, in addition to the relative geometry, is presented a concise overview as to how the elements of a three-way array can be transformed into compositional form and the relative geometry is given.

CoDa in three-way arrays and relative samples spaces

GALLO, Michele
2012-01-01

Abstract

The object of these short notes is to give a set of convenient symbols to define the sample space for the different compositional vectors that can be arranged into a three-way array. For the exploratory analysis of three-way data, Parafac/Candecomp and Tucker3 are some of the most applied models for low-rank decomposition of three-way arrays. Here, in addition to the relative geometry, is presented a concise overview as to how the elements of a three-way array can be transformed into compositional form and the relative geometry is given.
File in questo prodotto:
File Dimensione Formato  
12261-14135-1-PB (1).pdf

accesso aperto

Tipologia: Documento in Post-print
Licenza: PUBBLICO - Pubblico senza Copyright
Dimensione 1.98 MB
Formato Adobe PDF
1.98 MB Adobe PDF Visualizza/Apri

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/11574/40718
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
social impact