Gaussian graphical models play an important role in various areas such as genetics, finance, statistical physics and others. They are a powerful modelling tool, which allows one to describe the relationships among the variables of interest. From the Bayesian perspective, there are two sources of randomness: one is related to the multivariate distribution and the quantities that may parametrise the model, and the other has to do with the underlying graph, G, equivalent to describing the conditional independence structure of the model under consideration. In this paper, we propose a prior on G based on two loss components. One considers the loss in information one would incur in selecting the wrong graph, while the second penalises for large number of edges, favouring sparsity. We illustrate the prior on simulated data and on real datasets, and compare the results with other priors on G used in the literature. Moreover, we present a default choice of the prior as well as discuss how it can be calibrated so as to reflect available prior information.

A loss-based prior for Gaussian Graphical Models / L. Hinoveanu, F. Leisen, C. Villa. - In: AUSTRALIAN & NEW ZEALAND JOURNAL OF STATISTICS. - ISSN 1467-842X. - 62:4(2020), pp. 444-466. [10.1111/anzs.12307]

A loss-based prior for Gaussian Graphical Models

C. Villa
2020

Abstract

Gaussian graphical models play an important role in various areas such as genetics, finance, statistical physics and others. They are a powerful modelling tool, which allows one to describe the relationships among the variables of interest. From the Bayesian perspective, there are two sources of randomness: one is related to the multivariate distribution and the quantities that may parametrise the model, and the other has to do with the underlying graph, G, equivalent to describing the conditional independence structure of the model under consideration. In this paper, we propose a prior on G based on two loss components. One considers the loss in information one would incur in selecting the wrong graph, while the second penalises for large number of edges, favouring sparsity. We illustrate the prior on simulated data and on real datasets, and compare the results with other priors on G used in the literature. Moreover, we present a default choice of the prior as well as discuss how it can be calibrated so as to reflect available prior information.
Gaussian graphical models; information loss; loss‐ based prior; objective Bayes
Settore SECS-S/01 - Statistica
Article (author)
File in questo prodotto:
File Dimensione Formato  
1812.05531.pdf

accesso aperto

Tipologia: Pre-print (manoscritto inviato all'editore)
Dimensione 579.53 kB
Formato Adobe PDF
579.53 kB Adobe PDF Visualizza/Apri
anzs.12307.pdf

accesso riservato

Tipologia: Publisher's version/PDF
Dimensione 843.55 kB
Formato Adobe PDF
843.55 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

Caricamento pubblicazioni consigliate

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/2434/794792
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 1
social impact