The Annals of Statistics

Robust Regression: Asymptotics, Conjectures and Monte Carlo

Peter J. Huber

Full-text: Open access

Abstract

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.

Article information

Source
Ann. Statist. Volume 1, Number 5 (1973), 799-821.

Dates
First available in Project Euclid: 12 April 2007

Permanent link to this document
http://projecteuclid.org/euclid.aos/1176342503

Digital Object Identifier
doi:10.1214/aos/1176342503

Mathematical Reviews number (MathSciNet)
MR356373

Zentralblatt MATH identifier
0289.62033

JSTOR
links.jstor.org

Citation

Huber, Peter J. Robust Regression: Asymptotics, Conjectures and Monte Carlo. Ann. Statist. 1 (1973), no. 5, 799--821. doi:10.1214/aos/1176342503. http://projecteuclid.org/euclid.aos/1176342503.


Export citation