The bias of isotonic regression

DOI10.1214/20-EJS1677MaRDI QIDQ2293726zbMATH OpenOpenAlexWikidataFDO

Authors Ran Dai, Hyebin Song, Rina Foygel Barber, Garvesh Raskutti

Publication date 5 February 2020

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

Full work available at URL https://arxiv.org/abs/1908.04462, https://projecteuclid.org/euclid.ejs/1580871776

zbMATH Keywords

biasisotonic regression

Mathematics Subject Classification ID

Asymptotic distribution theory in statistics (62E20) Nonparametric estimation (62G05) Nonparametric regression and quantile regression (62G08) Asymptotic properties of nonparametric inference (62G20)

Abstract: We study the bias of the isotonic regression estimator. While there is extensive work characterizing the mean squared error of the isotonic regression estimator, relatively little is known about the bias. In this paper, we provide a sharp characterization, proving that the bias scales as $$ up to log factors, where $$ is the exponent corresponding to H{"o}lder smoothness of the underlying mean. Importantly, this result only requires a strictly monotone mean and that the noise distribution has subexponential tails, without relying on symmetric noise or other restrictive assumptions.

Recommendations

• Small sample properties of isotonic estimators - bias and mean squared error of linear combinations under normality
• Sharp asymptotics for isotonic regression
• Evaluation of the bias of the isotonic regression
• Risk bounds in isotonic regression
• Distribution-free properties of isotonic regression

Cites work

• A TEST OF HOMOGENEITY FOR ORDERED ALTERNATIVES
• A TEST OF HOMOGENEITY FOR ORDERED ALTERNATIVES. II
• Adaptive estimation of a quadratic functional by model selection.
• Adaptivity and optimality of the monotone least-squares estimator
• Contraction and uniform convergence of isotonic regression
• Divide and conquer in nonstandard problems and the super-efficiency phenomenon
• scientific article; zbMATH DE number 3126936 (Why is no real title available?)
• scientific article; zbMATH DE number 3949392 (Why is no real title available?)
• scientific article; zbMATH DE number 193111 (Why is no real title available?)
• scientific article; zbMATH DE number 3390139 (Why is no real title available?)
• scientific article; zbMATH DE number 3092910 (Why is no real title available?)
• Isotonic regression in general dimensions
• Maximum Likelihood Estimates of Monotone Parameters
• Nonparametric shape-restricted regression
• On estimation of isotonic piecewise constant signals
• On risk bounds in isotonic and other shape restricted regression problems
• Sharp asymptotics for isotonic regression
• The asymptotic behavior of monotone regression estimates
• THE COMPLETE AMALGAMATION INTO BLOCKS, BY WEIGHTED MEANS, OF A FINITE SET OF REAL NUMBERS
• The limit distribution of the \(L_{\infty}\)-error of Grenander-type estimators

Cited in

(4)

• Reconstruction of functions from prescribed proximal points
• Evaluation of the bias of the isotonic regression
• Small sample properties of isotonic estimators - bias and mean squared error of linear combinations under normality
• Consistent Estimation of Distribution Functions under Increasing Concave and Convex Stochastic Ordering

Describes a project that uses

Uses Software

• isotone

This page was built for publication: The bias of isotonic regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2293726)