论文信息 - Robust Bootstrap for S-estimators of Multivariate Regression

Robust Bootstrap for S-estimators of Multivariate Regression

Classical bootstrap applied to robust regression estimators can be extremely time consuming and the breakdown point of the procedure is lower than that of the estimator itself. In this paper we develop a robust bootstrap for S-estimators of multivariate regression. Through a simulation study it is shown that confidence intervals for the regression coefficients based on the robust bootstrap have good performance compared to other methods.

Stefan Van Aelst | G. Willems

[1] Arnold J. Stromberg,et al. Robust covariance estimates based on resampling , 1997 .

[2] P. L. Davies,et al. Asymptotic behaviour of S-estimates of multivariate location parameters and dispersion matrices , 1987 .

[3] Peter J. Rousseeuw,et al. Robust regression and outlier detection , 1987 .

[4] R. Zamar,et al. Bootstrapping robust estimates of regression , 2002 .

[5] B. Efron. Bootstrap Methods: Another Look at the Jackknife , 1979 .

[6] Anthony C. Davison,et al. Bootstrap Methods and Their Application , 1998 .

[7] Peter J. Rousseeuw,et al. ROBUST REGRESSION BY MEANS OF S-ESTIMATORS , 1984 .

[8] D. Ruppert. Computing S Estimators for Regression and Multivariate Location/Dispersion , 1992 .

[9] H. P. Lopuhaä. On the relation between S-estimators and M-estimators of multivariate location and covariance , 1989 .

[10] W. Härdle,et al. Robust and Nonlinear Time Series Analysis , 1984 .

[11] Pierre Duchesne,et al. Robust estimation of the SUR model , 2000 .

[12] Kesar Singh,et al. Breakdown theory for bootstrap quantiles , 1998 .

[13] Richard A. Johnson,et al. Applied Multivariate Statistical Analysis , 1983 .