User: Guest  Login
Title:

A Bayesian information criterion for singular models

Document type:
Zeitschriftenaufsatz
Author(s):
Drton, Mathias; Plummer, Martyn
Abstract:
We consider approximate Bayesian model choice for model selection problems that involve models whose Fisher information matrices may fail to be invertible along other competing submodels. Such singular models do not obey the regularity conditions underlying the derivation of Schwarz's Bayesian information criterion BIC and the penalty structure in BIC generally does not reflect the frequentist large sample behaviour of the marginal likelihood. Although large sample theory for the marginal likeli...     »
Keywords:
Bayesian information criterion, Factor analysis, Mixture model, Model selection, Reduced rank regression, Schwarz information criterion, Singular learning theory
Dewey Decimal Classification:
510 Mathematik
Journal title:
Journal of the Royal Statistical Society: Series B (Statistical Methodology)
Year:
2017
Journal volume:
79
Year / month:
2017-02
Quarter:
1. Quartal
Month:
Feb
Journal issue:
2
Pages contribution:
323-380
Language:
en
Fulltext / DOI:
doi:10.1111/rssb.12187
Publisher:
Wiley
E-ISSN:
1369-7412
Status:
Verlagsversion / published
Date of publication:
07.02.2017
 BibTeX