A note on the approximate admissibility of regularized estimators in the Gaussian sequence model

DOI10.1214/17-EJS1354MaRDI QIDQ1684161zbMATH OpenOpenAlexFDO

Authors Xi Chen, Adityanand Guntuboyina, Yuchen Zhang

Publication date 8 December 2017

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1703.00542, https://projecteuclid.org/euclid.ejs/1511492461

zbMATH Keywords

admissibilityBayes riskGaussian sequence modelleast squares estimatorminimaxity

Mathematics Subject Classification ID

Admissibility in statistical decision theory (62C15) Minimax procedures in statistical decision theory (62C20) Parametric inference under constraints (62F30) Nonparametric estimation (62G05) Estimation in multivariate analysis (62H12)

Abstract: We study the problem of estimating an unknown vector $heta$ from an observation $X$ drawn according to the normal distribution with mean $heta$ and identity covariance matrix under the knowledge that $heta$ belongs to a known closed convex set $Theta$. In this general setting, Chatterjee (2014) proved that the natural constrained least squares estimator is "approximately admissible" for every $Theta$. We extend this result by proving that the same property holds for all convex penalized estimators as well. Moreover, we simplify and shorten the original proof considerably. We also provide explicit upper and lower bounds for the universal constant underlying the notion of approximate admissibility.

Recommendations

• A new perspective on least squares under convex constraint
• Adaptive and efficient estimation in the Gaussian sequence model
• Penalized blockwise Stein's method, monotone oracles and sharp adaptive estimation
• Adaptive estimation of a quadratic functional by model selection.
• Universal inadmissibility of least squares estimator

Cited in

(2)

• Nonparametric shape-restricted regression
• A new perspective on least squares under convex constraint

This page was built for publication: A note on the approximate admissibility of regularized estimators in the Gaussian sequence model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1684161)