A Minimax-Bias Property of the Least $\alpha$-Quantile Estimates
Yohai, Victor J. ; Zamar, Ruben H.
Ann. Statist., Tome 21 (1993) no. 1, p. 1824-1842 / Harvested from Project Euclid
A natural measure of the degree of robustness of an estimate $\mathbf{T}$ is the maximum asymptotic bias $B_\mathbf{T}(\varepsilon)$ over an $\varepsilon$-contamination neighborhood. Martin, Yohai and Zamar have shown that the class of least $\alpha$-quantile regression estimates is minimax bias in the class of $M$-estimates, that is, they minimize $B_\mathbf{T}(\varepsilon)$, with $\alpha$ depending on $\varepsilon$. In this paper we generalize this result, proving that the least $\alpha$-quantile estimates are minimax bias in a much broader class of estimates which we call residual admissible and which includes most of the known robust estimates defined as a function of the regression residuals (e.g., least median of squares, least trimmed of squares, $S$-estimates, $\tau$-estimates, $M$-estimates, signed $R$-estimates, etc.). The minimax results obtained here, likewise the results obtained by Martin, Yohai and Zamar, require that the carriers have elliptical distribution under the central model.
Publié le : 1993-12-14
Classification:  Minimax bias,  regression,  robust estimates,  62F35
@article{1176349400,
     author = {Yohai, Victor J. and Zamar, Ruben H.},
     title = {A Minimax-Bias Property of the Least $\alpha$-Quantile Estimates},
     journal = {Ann. Statist.},
     volume = {21},
     number = {1},
     year = {1993},
     pages = { 1824-1842},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176349400}
}
Yohai, Victor J.; Zamar, Ruben H. A Minimax-Bias Property of the Least $\alpha$-Quantile Estimates. Ann. Statist., Tome 21 (1993) no. 1, pp.  1824-1842. http://gdmltest.u-ga.fr/item/1176349400/