Oracle inequalities for square root analysis estimators with application to total variation penalties

DOI10.1093/IMAIAI/IAAA002MaRDI QIDQ5155953zbMATH OpenOpenAlexFDO

Authors Francesco Ortelli, Sara Van De Geer

Publication date 13 October 2021

Published in Information and Inference: A Journal of the IMA (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1902.11192

zbMATH Keywords

analysiscycle graphedge LassoLassonullspaceoracle inequalitysparsitysquare root Lassototal variation regularizationtrend filtering

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Ridge regression; shrinkage estimators (Lasso) (62J07) Inference from stochastic processes and prediction (62M20)

Abstract: Through the direct study of the analysis estimator we derive oracle inequalities with fast and slow rates by adapting the arguments involving projections by Dalalyan, Hebiri and Lederer (2017). We then extend the theory to the square root analysis estimator. Finally, we focus on (square root) total variation regularized estimators on graphs and obtain constant-friendly rates, which, up to log-terms, match previous results obtained by entropy calculations. We also obtain an oracle inequality for the (square root) total variation regularized estimator over the cycle graph.

Recommendations

• Oracle inequalities and nonparametric function estimation
• On the strong oracle property of concave penalized estimators with infinite penalty derivative at the origin
• Sharp oracle inequalities for square root regularization
• From Ordinary to Shrinkage Square-Root Estimators
• Oracle inequalities for probability density estimations
• Sharp oracle inequalities for least squares estimators in shape restricted regression
• Sharp Oracle Inequalities for Stationary Points of Nonconvex Penalized M-Estimators
• Asymptotic oracle properties of SCAD-penalized least squares estimators
• Sharp oracle inequalities for aggregation of affine estimators
• Oracle inequalities for Efromovich--Pinsker blockwise estimates

Cited in

(4)

• A sharp oracle inequality for graph-slope
• Prediction bounds for higher order total variation regularized least squares
• Tensor denoising with trend filtering
• A cross-validation framework for signal denoising with applications to trend filtering, dyadic CART and beyond

This page was built for publication: Oracle inequalities for square root analysis estimators with application to total variation penalties

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5155953)