On multivariate median regression

DOI10.2307/3318697MaRDI QIDQ1304027zbMATH OpenOpenAlexFDO

Authors B. Chakraborty

Publication date 30 August 2000

Published in Bernoulli (Search for Journal in Brave)

Full work available at URL https://projecteuclid.org/euclid.bj/1171899324

zbMATH Keywords

affine equivariancebootstrapefficiencyelliptically symmetric distributionsgeneralized varianceleast absolute deviationsmultiresponse linear modelstandard error estimationtransformation-retransformation estimate

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Linear regression; mixed models (62J05)

Review text

The author considers an extension of the concept of least absolute deviation regression for problems with multivariate response. The approach is based on a transformation and retransformation technique that chooses a data-driven coordinate system for transforming the response vectors and then retransform the estimate of the matrix of regression parameters, which is obtained by performing coordinatewise least absolute deviations regression on the transformed response vectors. It is shown that the estimates are equivariant under non-singular linear transformations of the response vectors. An algorithm, called TREMMER (Transformation Retransformation Estimates in Multivariate Median Regression), is suggested, which adaptively chooses the optimal data-driven coordinate system and then computes the regression estimates. Resampling techniques like the bootstrap are used to conveniently estimate the standard errors of TREMMER estimates. The proposed estimate is more efficient than the non-equivariant coordinatewise least absolute deviations estimate, and it outperforms ordinary least-squares estimates in the case of heavy-tailed non-normal multivariate error distributions. Asymptotic normality and some other optimality properties of the adaptive transformation-retransformation median estimate are also discussed. The performance of the procedure is demonstrated with two real examples.

Recommendations

• On multivariate quantile regression
• On an Adaptive Transformation–Retransformation Estimate of Multivariate Location
• Least Median of Squares Regression
• The multivariate least-trimmed squares estimator

Cited in

(21)

• Multivariate generalized S-estimators
• Permutation-based multivariate regression analysis: The case for least sum of absolute deviations regression
• On multivariate quantile regression
• Bayesian multivariate quantile regression using dependent Dirichlet process prior
• Gibbs posterior inference on multivariate quantiles
• Regularized multiple mediation analysis
• A multivariate Wilcoxon regression estimate
• TR Multivariate Conditional Median Estimation
• scientific article; zbMATH DE number 5280064 (Why is no real title available?)
• On Adaptive Transformation–Retransformation Estimate of Conditional Spatial Median
• The multivariate L ₁ -median and associated data depth
• Estimates of Regression Coefficients Based on the Sign Covariance Matrix
• Robust estimation of coefficient matrices in multivariate linear regression models
• On the statistical efficiency of robust estimators of multivariate location
• scientific article; zbMATH DE number 2210186 (Why is no real title available?)
• Robust inference theory for non-regular time series models and its extensions
• The application of multiple-output quantile regression to the US financial cycle
• Multivariate medial correlation with applications
• Multivariate least-trimmed squares regression estimator
• Bootstrapping least distance estimator in the multivariate regression model
• Robust online signal extraction from multivariate time series

This page was built for publication: On multivariate median regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1304027)