In this short paper, we discuss a novel way of constructing prior distributions for correlation matrices and an associated approach to inference. We construct a prior penalizing large correlations, which we incorporate into an oblique factor model and a Candecomp/Parafac model for three-way data. We argue that this choice of prior for the factor correlation matrix, combined with a shrinkage prior for elements of the factor loadings matrix, leads to interpretable solutions. At the meeting we will demonstrate this through applications to real data.
A Bayesian oblique factor model with extension to tensor data / Jauch, M.; Giordani, P.; Dunson, D.. - (2017), pp. 553-560. (Intervento presentato al convegno SIS 2017 tenutosi a Firenze).
A Bayesian oblique factor model with extension to tensor data
Giordani P.;
2017
Abstract
In this short paper, we discuss a novel way of constructing prior distributions for correlation matrices and an associated approach to inference. We construct a prior penalizing large correlations, which we incorporate into an oblique factor model and a Candecomp/Parafac model for three-way data. We argue that this choice of prior for the factor correlation matrix, combined with a shrinkage prior for elements of the factor loadings matrix, leads to interpretable solutions. At the meeting we will demonstrate this through applications to real data.File | Dimensione | Formato | |
---|---|---|---|
Giordani_Bayesian-oblique-factor_2017.pdf
accesso aperto
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Creative commons
Dimensione
1.67 MB
Formato
Adobe PDF
|
1.67 MB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.