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.