liu.seSearch for publications in DiVA
Change search
ReferencesLink to record
Permanent link

Direct link
Parameter estimation and order selection for linear regressionproblems
Dept. of Information Technology, Uppsala University, Sweden.
KTH/EE Communication Theory Laboratory, Royal Institute of Technology, Sweden.ORCID iD: 0000-0002-7599-4367
2006 (English)In: Proc. of European Signal Processing Conference (EUSIPCO), Sept. 2006, 2006Conference paper (Refereed)
Abstract [en]

Parameter estimation and model order selection for linear regression models are two classical problems. In this article we derive the minimum mean-square error (MMSE) parameter estimate for a linear regressionmodel with unknown order. We call the so-obtained estimator the Bayesian Parameter estimation Method (BPM). We also derive the model order selection rule which maximizes the probability of selecting the correct model. The rule is denoted BOSS—Bayesian Order Selection Strategy. The estimators have several advantages: They satisfy certain optimality criteria, they are non-asymptotic and they have low computational complexity. We also derive “empirical Bayesian” versions of BPM and BOSS, which do not require any prior knowledge nor do they need the choice of any “user parameters”. We show that our estimators outperform several classical methods, including the AIC and BIC for orderselection.

Place, publisher, year, edition, pages
National Category
Engineering and Technology
URN: urn:nbn:se:liu:diva-77238OAI: diva2:525710
14th European Signal Processing Conference (EUSIPCO 2006), Florence, Italy, September 4-8, 2006
Available from: 2012-05-09 Created: 2012-05-09 Last updated: 2016-08-31Bibliographically approved

Open Access in DiVA

No full text

Other links

Parameter estimation and order selection for linear regression problems

Search in DiVA

By author/editor
Larsson, Erik G.
Engineering and Technology

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

Total: 104 hits
ReferencesLink to record
Permanent link

Direct link