Nearly root-n approximation for regression quantile processes

DOI10.1214/12-AOS1021zbMATH Open1284.62291arXiv1210.1092OpenAlexW2081314317MaRDI QIDQ693744FDOQ693744

Authors: Stephen Portnoy 

Publication date: 10 December 2012

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: Traditionally, assessing the accuracy of inference based on regression quantiles has relied on the Bahadur representation. This provides an error of order $n^{-1/4}$ in normal approximations, and suggests that inference based on regression quantiles may not be as reliable as that based on other (smoother) approaches, whose errors are generally of order $n^{-1/2}$ (or better in special symmetric cases). Fortunately, extensive simulations and empirical applications show that inference for regression quantiles shares the smaller error rates of other procedures. In fact, the "Hungarian" construction of Koml'{o}s, Major and Tusn'{a}dy [Z. Wahrsch. Verw. Gebiete 32 (1975) 111-131, Z. Wahrsch. Verw. Gebiete 34 (1976) 33-58] provides an alternative expansion for the one-sample quantile process with nearly the root-$n$ error rate (specifically, to within a factor of $logn$). Such an expansion is developed here to provide a theoretical foundation for more accurate approximations for inference in regression quantile models. One specific application of independent interest is a result establishing that for conditional inference, the error rate for coverage probabilities using the Hall and Sheather [J. R. Stat. Soc. Ser. B Stat. Methodol. 50 (1988) 381-391] method of sparsity estimation matches their one-sample rate.

Full work available at URL: https://arxiv.org/abs/1210.1092

Recommendations

• scientific article; zbMATH DE number 2143284
• Nonstandard quantile-regression inference
• Goodness of Fit and Related Inference Processes for Quantile Regression
• Saddlepoint tests for quantile regression
• On nonparametric regression estimators based on regression quantiles

zbMATH Keywords

regression quantilesasymptotic approximationHungarian construction

Mathematics Subject Classification ID

Asymptotic properties of nonparametric inference (62G20) Linear inference, regression (62J99) Order statistics; empirical distribution functions (62G30) Functional limit theorems; invariance principles (60F17)

Cites Work

• Regression Quantiles
• A resampling method based on pivotal estimating functions
• Quantile regression.
• Title not available (Why is that?)
• An approximation of partial sums of independent RV's, and the sample DF. II
• An approximation of partial sums of independent RV'-s, and the sample DF. I
• Probability Inequalities for Sums of Bounded Random Variables
• Markov Chain Marginal Bootstrap
• Title not available (Why is that?)
• Bootstrap Methods for Median Regression Models
• Tail Probability Approximations
• Extensions of results of Komlós, Major, and Tusnády to the multivariate case
• Tests of linear hypotheses based on regression rank scores
• Direct use of regression quantiles to construct confidence sets in linear models
• On monotonicity of regression quantile functions
• Analytical and Bootstrap Approximations to Estimator Distributions in L 1 Regression

Cited In (9)

• The second-order bias of quantile estimators
• Saddlepoint tests for quantile regression
• On general notions of depth for regression
• Quantile-Regression Inference With Adaptive Control of Size
• Smoothed quantile regression with large-scale inference
• A conversation with Stephen Portnoy
• Bahadur representations for bootstrap quantiles
• A direct approach to inference in nonparametric and semiparametric quantile models
• Smoothing Quantile Regressions

Uses Software

• quantreg

This page was built for publication: Nearly root-\(n\) approximation for regression quantile processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q693744)