Least sum of squares of trimmed residuals regression

DOI10.1214/23-EJS2164MaRDI QIDQ6184883zbMATH OpenOpenAlexFDO

Authors Yijun Zuo, Hanwen Zuo

Publication date 5 January 2024

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

Full work available at URL https://arxiv.org/abs/2202.10329, https://projecteuclid.org/journals/electronic-journal-of-statistics/volume-17/issue-2/Least-sum-of-squares-of-trimmed-residuals-regression/10.1214/23-EJS2164.full

zbMATH Keywords

consistencyrobust regressionfinite sample breakdown pointtrimmed residualsapproximate computation algorithm

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Linear inference, regression (62J99) Nonparametric inference (62G99)

Abstract: In the famous least sum of trimmed squares (LTS) of residuals estimator (Rousseeuw (1984)), residuals are first squared and then trimmed. In this article, we first trim residuals - using a depth trimming scheme - and then square the rest of residuals. The estimator that can minimize the sum of squares of the trimmed residuals, is called an LST estimator. It turns out that LST is a robust alternative to the classic least sum of squares (LS) estimator. Indeed, it has a very high finite sample breakdown point, and can resist, asymptotically, up to $50\mathrm{\%}$ contamination without breakdown - in sharp contrast to the $0\mathrm{\%}$ of the LS estimator. The population version of LST is Fisher consistent, and the sample version is strong and root-$n$ consistent and asymptotically normal. Approximate algorithms for computing LST are proposed and tested in synthetic and real data examples. These experiments indicate that one of the algorithms can compute the LST estimator very fast and with relatively smaller variances than the famous LTS estimator. All the evidence suggests that LST deserves to be a robust alternative to the LS estimator and is feasible in practice for high dimensional data sets (with possible contamination and outliers).

Cites work

• scientific article; zbMATH DE number 3842984 (Why is no real title available?)
• scientific article; zbMATH DE number 3829050 (Why is no real title available?)
• scientific article; zbMATH DE number 3905646 (Why is no real title available?)
• scientific article; zbMATH DE number 3954047 (Why is no real title available?)
• scientific article; zbMATH DE number 2058049 (Why is no real title available?)
• scientific article; zbMATH DE number 194744 (Why is no real title available?)
• scientific article; zbMATH DE number 2107836 (Why is no real title available?)
• scientific article; zbMATH DE number 3188867 (Why is no real title available?)
• A new approach for the computation of halfspace depth in high dimensions
• Approximation Theorems of Mathematical Statistics
• Asymptotic Statistics
• Breakdown properties of location estimates based on halfspace depth and projected outlyingness
• Computation of projection regression depth and its induced median
• Convergence of stochastic processes
• Cube root asymptotics
• General notions of statistical depth function.
• Hedonic housing prices and the demand for clean air
• High Breakdown-Point Estimates of Regression by Means of the Minimization of an Efficient Scale
• High breakdown-point and high efficiency robust estimates for regression
• Improved feasible solution algorithms for high breakdown estimation.
• Large sample properties of the regression depth induced median
• Least Median of Squares Regression
• Maximal inequalities for degenerate U-processes with applications to optimization estimators
• Multidimensional trimming based on projection depth
• On general notions of depth for regression
• One-Step Huber Estimates in the Linear Model
• Projection-based depth functions and associated medians
• Regression Depth
• Robust Estimation of a Location Parameter
• Robust Statistics
• Robust regression: Asymptotics, conjectures and Monte Carlo
• Robustness of the deepest projection regression functional
• The Least Trimmed Differences Regression Estimator and Alternatives
• The Limiting Distribution of the Maximum Rank Correlation Estimator
• The anonymous professor Gergonne
• The feasible solution algorithm for least trimmed squares regression
• The influence function of penalized regression estimators
• The influence functions for the least trimmed squares and the least trimmed absolute deviations estimators
• The least trimmed squares. I: Consistency.
• The least trimmed squares. II: \(\sqrt {n}\)-consistency
• The least trimmed squares. III: Asymptotic normality.
• The trimmed mean in the linear model
• Trimmed Least Squares Estimation in the Linear Model
• Trimmed and Winsorized means based on a scaled deviation
• Weak convergence and empirical processes. With applications to statistics

Cited in

(4)

• Multivariate least-trimmed squares regression estimator
• Trimmed slope estimates for simple linear regression
• Least squares residuals and minimal residual methods
• scientific article; zbMATH DE number 3992657 (Why is no real title available?)

This page was built for publication: Least sum of squares of trimmed residuals regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6184883)