Weakly decomposable regularization penalties and structured sparsity

type="main" xml:id="sjos12032-abs-0001"> <title type="main">ABSTRACT</title>It has been shown in literature that the Lasso estimator, or ℓ<sub>1</sub>-penalized least squares estimator, enjoys good oracle properties. This paper examines which special properties of the ℓ<sub>1</sub>-penalty allow for sharp oracle results, and then extends the situation to general norm-based penalties that satisfy a weak decomposability condition.

MoreLess

Year of publication:	2014
Authors:	Geer, Sara
Published in:	Scandinavian Journal of Statistics. - Danish Society for Theoretical Statistics, ISSN 0303-6898. - Vol. 41.2014, 1, p. 72-86
Publisher:	Danish Society for Theoretical Statistics Finnish Statistical Society Norwegian Statistical Association Swedish Statistical Association

More details

Extent:	text/html
Type of publication:	Article
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10011153109