High breakdown point robust regression with censored data
Salibian-Barrera, Matías ; Yohai, Víctor J.
Ann. Statist., Tome 36 (2008) no. 1, p. 118-146 / Harvested from Project Euclid
In this paper, we propose a class of high breakdown point estimators for the linear regression model when the response variable contains censored observations. These estimators are robust against high-leverage outliers and they generalize the LMS (least median of squares), S, MM and τ-estimators for linear regression. An important contribution of this paper is that we can define consistent estimators using a bounded loss function (or equivalently, a redescending score function). Since the calculation of these estimators can be computationally costly, we propose an efficient algorithm to compute them. We illustrate their use on an example and present simulation studies that show that these estimators also have good finite sample properties.
Publié le : 2008-02-15
Classification:  Robust estimates,  censored data,  high breakdown point estimates,  linear regression model,  accelerated failure time models,  62F35,  62J05
@article{1201877296,
     author = {Salibian-Barrera, Mat\'\i as and Yohai, V\'\i ctor J.},
     title = {High breakdown point robust regression with censored data},
     journal = {Ann. Statist.},
     volume = {36},
     number = {1},
     year = {2008},
     pages = { 118-146},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1201877296}
}
Salibian-Barrera, Matías; Yohai, Víctor J. High breakdown point robust regression with censored data. Ann. Statist., Tome 36 (2008) no. 1, pp.  118-146. http://gdmltest.u-ga.fr/item/1201877296/