Robust Regression: Asymptotics, Conjectures and Monte Carlo
Huber, Peter J.
Ann. Statist., Tome 1 (1973) no. 2, p. 799-821 / Harvested from Project Euclid
Text on Project Euclid
PDF
Alt PDF
Maximum likelihood type robust estimates of regression are defined and their asymptotic properties are investigated both theoretically and empirically. Perhaps the most important new feature is that the number $p$ of parameters is allowed to increase with the number $n$ of observations. The initial terms of a formal power series expansion (essentially in powers of $p/n$) show an excellent agreement with Monte Carlo results, in most cases down to 4 observations per parameter.
Publié le : 1973-09-14
Classification:  
@article{1176342503,
     author = {Huber, Peter J.},
     title = {Robust Regression: Asymptotics, Conjectures and Monte Carlo},
     journal = {Ann. Statist.},
     volume = {1},
     number = {2},
     year = {1973},
     pages = { 799-821},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176342503}
}

Huber, Peter J. Robust Regression: Asymptotics, Conjectures and Monte Carlo. Ann. Statist., Tome 1 (1973) no. 2, pp.  799-821. http://gdmltest.u-ga.fr/item/1176342503/