One-step R-estimation in linear models with stable errors

DOI10.1080/10485252.2010.525234zbMATH Open1443.62058arXiv1210.5073OpenAlexW2144988112MaRDI QIDQ528136FDOQ528136

Yvik Swan, David Veredas, Thomas Verdebout, Marc Hallin

Publication date: 12 May 2017

Published in: Journal of Econometrics, Journal of Nonparametric Statistics (Search for Journal in Brave)

Abstract: Classical estimation techniques for linear models either are inconsistent, or perform rather poorly, under $alpha$-stable error densities; most of them are not even rate-optimal. In this paper, we propose an original one-step R-estimation method and investigate its asymptotic performances under stable densities. Contrary to traditional least squares, the proposed R-estimators remain root-$n$ consistent (the optimal rate) under the whole family of stable distributions, irrespective of their asymmetry and tail index. While parametric stable-likelihood estimation, due to the absence of a closed form for stable densities, is quite cumbersome, our method allows us to construct estimators reaching the parametric efficiency bounds associated with any prescribed values $(alph{a}_0,b_0)$ of the tail index $alpha$ and skewness parameter $b$, while preserving root-$n$ consistency under any $(alpha,b)$ as well as under usual light-tailed densities. The method furthermore avoids all forms of multidimensional argmin computation. Simulations confirm its excellent finite-sample performances.

Full work available at URL: https://arxiv.org/abs/1210.5073

zbMATH Keywords

rank testsstable distributionsasymptotic relative efficiencylocal asymptotic normalityLAD estimationR-estimationasymptotic relative efficiencies

Mathematics Subject Classification ID

Infinitely divisible distributions; stable distributions (60E07) Point estimation (62F10) Asymptotic properties of parametric estimators (62F12) Nonparametric hypothesis testing (62G10) Applications of statistics to economics (62P20) Asymptotic distribution theory in statistics (62E20) Stable stochastic processes (60G52)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Asymptotic Statistics
• The method of simulated quantiles
• Testing Statistical Hypotheses
• Nonparametric Confidence Intervals for a Shift Parameter
• Estimates of Location Based on Rank Tests
• A Simplex Method for Function Minimization
• Asymptotics in statistics. Some basic concepts.
• Semiparametrically efficient rank-based inference for shape. II: Optimal \(R\)-estimation of shape
• Asymptotic methods in statistical decision theory
• Asymptotic Theory of Least Absolute Error Regression
• Numerical calculation of stable densities and distribution functions
• Nonparametric Estimate of Regression Coefficients
• Asymptotics in statistics: some basic concepts
• Weighted empirical processes in dynamic nonlinear models.
• On the asymptotic normality of the maximum-likelihood estimate when sampling from a stable distribution
• Asymptotic Linearity of a Rank Statistic in Regression Parameter
• Parameter Estimates for Symmetric Stable Distributions
• When is the Student \(t\)-statistic asymptotically standard normal?
• Limit distributions of self-normalized sums
• Limiting distributions of the non-central \(t\)-statistic and their applications to the power of \(t\)-tests under non-normality
• Student's t-Test Under Symmetry Conditions
• Portfolio Analysis in a Stable Paretian Market
• Maximum likelihood estimation for \(\alpha \)-stable autoregressive processes
• Asymptotic Behavior of a Class of Confidence Regions Based on Ranks in Regression
• Simple consistent estimators of stable distribution parameters
• Estimation in Univariate and Multivariate Stable Distributions
• Statistical decision theory. Estimation, testing, and selection.
• An algorithm for evaluating stable densities in Zolotarev's \((M)\) parameterization
• Semi-parametric efficiency, distribution-freeness and invariance
• Stable Distributions in Statistical Inference: 2. Information from Stably Distributed Samples
• Bootstrapping sample quantiles in non-regular cases
• One-step R-estimation in linear models with stable errors
• \(R\)-estimation of the parameters of autoregressive [AR(\(p\))] models
• R-estimation for arma models
• \(L_1\)-estimation in linear models with heterogeneous white noise
• Linear models without moments
• On a Distribution-free Method of Estimating Asymptotic Efficiency of a Class of Non-parametric Tests

Cited In (25)

• A class of optimization problems motivated by rank estimators in robust regression
• A note on the normal approximation error for randomly weighted self-normalized sums
• Statistical modeling of the Cobb-Douglas production function: a multiple linear regression approach in presence of stable distribution noise
• Optimal rank-based testing for principal components
• Rank tests for corrupted linear models
• Title not available (Why is that?)
• GTL regression: a linear model with skewed and thick-tailed disturbances
• On Hodges and Lehmann’s “6/π Result”
• Aligned rank tests for the linear model with heteroscedastic errors
• Bayesian analysis of multivariate stable distributions using one-dimensional projections
• One-step R-estimation in linear models with stable errors
• Simultaneous inference based on rank statistics in linear models
• On linear rank tests for truncated binomial randomization
• Rank-Based Estimation and Associated Inferences for Linear Models With Cluster Correlated Errors
• Efficient and adaptive rank-based fits for linear models with skew-normal errors
• R-estimation in semiparametric dynamic location-scale models
• Statistical Tests and Estimators of the Rank of a Matrix and Their Applications in Econometric Modelling
• \(L_2\) differentiability of generalized linear models
• Title not available (Why is that?)
• A class of simple distribution-free rank-based unit root tests
• Modification of Malinowski’s F-test for pseudo rank estimation revisited
• Optimal rank-based tests against first-order superdiagonal bilinear dependence
• Measure of location-based estimators in simple linear regression
• Heavy tails of OLS
• Linear and nonlinear regression with stable errors

Uses Software

• STABLE

This page was built for publication: One-step R-estimation in linear models with stable errors

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q528136)