Author： Abramovich Felix Grinshtein Vadim
Publisher： Oxford University Press
ISSN： 1464-3510
Source： Biometrika, Vol.100, Iss.2, 2013-06, pp. : 355-370
Disclaimer: Any content in publications that violate the sovereignty, the constitution or regulations of the PRC is not accepted or approved by CNPIEC.
Abstract
We consider estimating a sparse group of sparse normal mean vectors, based on penalized likelihood estimation with complexity penalties on the number of nonzero mean vectors and the numbers of their significant components, which can be performed by a fast algorithm. The resulting estimators are developed within a Bayesian framework and can be viewed as maximum a posteriori estimators. We establish their adaptive minimaxity over a wide range of sparse and dense settings. A simulation study demonstrates the efficiency of the proposed approach, which successfully competes with the sparse group lasso estimator.
Related content
Scaled sparse linear regression
Biometrika, Vol. 99, Iss. 4, 2012-12 ,pp. :
Access to resources
Recommend
Is sparse and distributed the coding goal of simple cells?
By Zhao Li
Biological Cybernetics, Vol. 91, Iss. 6, 2004-12 ,pp. :