Linear Regression With a Sparse Parameter Vector
2007 (English)In: IEEE Transactions on Signal Processing, ISSN 1053-587X, E-ISSN 1941-0476, Vol. 55, no 2, 451-460 p.Article in journal (Refereed) Published
We consider linear regression under a model where the parameter vector is known to be sparse. Using a Bayesian framework, we derive the minimum mean-square error (MMSE) estimate of the parameter vector, and a computationally efficient approximation of it. We also derive an empirical-Bayesian version of the estimator, which does not need any a priori information, nor does it need the selection of any user parameters. As a byproduct, we obtain a powerful model (``basis'') selection tool for sparse models. The performance and robustness of our new estimators are illustrated via numerical examples.
Place, publisher, year, edition, pages
2007. Vol. 55, no 2, 451-460 p.
Linear regression, sparse models, Bayesian inference, MMSE estimation, basis selection, model selection, variable selection, model averaging, Lasso
Engineering and Technology
IdentifiersURN: urn:nbn:se:uu:diva-82080DOI: 10.1109/TSP.2006.887109ISI: 000243952600005OAI: oai:DiVA.org:uu-82080DiVA: diva2:109995