A simple and tractable iterative least squares estimation procedure for censored regression models with known error distributions is analyzed. It is found to be equivalent to a well-defined Huber type $M$-estimate. Under a regularity condition, the algorithm converges geometrically to a unique solution. The resulting estimate is $\sqrt N$-consistent and asymptotically normal.
Publié le : 1993-12-14
Classification:
Censored data,
iterative least squares,
geometrical convergence,
62F10,
62F12
@article{1176349394,
author = {Breiman, Leo and Tsur, Yacov and Zemel, Amos},
title = {On a Simple Estimation Procedure for Censored Regression Models with Known Error Distributions},
journal = {Ann. Statist.},
volume = {21},
number = {1},
year = {1993},
pages = { 1711-1720},
language = {en},
url = {http://dml.mathdoc.fr/item/1176349394}
}
Breiman, Leo; Tsur, Yacov; Zemel, Amos. On a Simple Estimation Procedure for Censored Regression Models with Known Error Distributions. Ann. Statist., Tome 21 (1993) no. 1, pp. 1711-1720. http://gdmltest.u-ga.fr/item/1176349394/