In the last years, there has been a growing interest in the analysis of three- way (matrix-variate) data via mixture models. Quite often real data are affected by outliers,andthisalsooccursinthecontextofthree-waydata.Acommonsolutionfor managing this type of data consists in fitting mixtures of heavy-tailed distributions. Unfortunately, the three-way literature is still limited, with the only matrix variate 푡 mixtures recently proposed to cope with this issue. Therefore, in this work we firstly introduceanewheavy-tailedmatrix-variatedistributionandthenweuseitwithinthe mixture model setting. The resulting mixture model is finally fitted to a real dataset for illustrative purposes.

A new mixture model for three-way data

Tomarchio S. D.
;
Punzo A.;Bagnato L.
2020-01-01

Abstract

In the last years, there has been a growing interest in the analysis of three- way (matrix-variate) data via mixture models. Quite often real data are affected by outliers,andthisalsooccursinthecontextofthree-waydata.Acommonsolutionfor managing this type of data consists in fitting mixtures of heavy-tailed distributions. Unfortunately, the three-way literature is still limited, with the only matrix variate 푡 mixtures recently proposed to cope with this issue. Therefore, in this work we firstly introduceanewheavy-tailedmatrix-variatedistributionandthenweuseitwithinthe mixture model setting. The resulting mixture model is finally fitted to a real dataset for illustrative purposes.
2020
9788891910776
Three-way data
Model-based clustering
Heavy-tailed distributions
File in questo prodotto:
File Dimensione Formato  
2020 - Tomarchio & Punzo - SIS Pisa.pdf

solo gestori archivio

Descrizione: Articolo principale
Tipologia: Versione Editoriale (PDF)
Dimensione 336.17 kB
Formato Adobe PDF
336.17 kB 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/20.500.11769/495576
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact