Change search
ReferencesLink to record
Permanent link

Direct link
Empirical Bayes linear regression with unknown model order
KTH, School of Electrical Engineering (EES), Communication Theory.
2008 (English)In: Digital signal processing (Print), ISSN 1051-2004, E-ISSN 1095-4333, Vol. 18, no 2, 236-248 p.Article in journal (Refereed) Published
Abstract [en]

We study maximum a posteriori probability model order selection for linear regression models, assuming Gaussian distributed noise and coefficient vectors. For the same data model, we also derive the minimum mean-square error coefficient vector estimate. The approaches are denoted BOSS (Bayesian order selection strategy) and BPM (Bayesian parameter estimation method), respectively. In their simplest form, both BOSS and BPM require a priori knowledge of the distribution of the coefficients. However, under the assumption that the coefficient variance profile is smooth, we derive "empirical Bayesian" versions of our algorithms which estimate the coefficient variance profile from the observations and thus require little or no information from the user. We show in numerical examples that the estimators can outperform several classical methods, including the well-known AICc and BIC for model order selection.

Place, publisher, year, edition, pages
2008. Vol. 18, no 2, 236-248 p.
Keyword [en]
linear regression, empirical Bayes MMSE estimation, parameter estimation, order selection
National Category
Computer Science
URN: urn:nbn:se:kth:diva-38118DOI: 10.1016/j.dsp.2007.03.005ISI: 000254781300013ScopusID: 2-s2.0-40049090332OAI: diva2:435955
14th European Signal Processing Conference Location: Florence, ITALY Date: SEP 04-08, 2006Available from: 2011-08-22 Created: 2011-08-22 Last updated: 2011-08-22Bibliographically approved

Open Access in DiVA

No full text

Other links

Publisher's full textScopus

Search in DiVA

By author/editor
Larsson, Erik G.
By organisation
Communication Theory
In the same journal
Digital signal processing (Print)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Altmetric score

Total: 13 hits
ReferencesLink to record
Permanent link

Direct link