We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
M-Estimates of regression when the scale is unknown and the error distribution is possibly asymmetric: A minimax result.
- Authors
Li, Bing; Zamar, Ruben H.
- Abstract
The article discusses a study on the minimax-variance approach pioneered by P.J. Huber in the local model when the scale parameter is known and the contamination distribution F is symmetric. It introduces a new robust regression estimate ofßn obtained by minimizing the maximum trace where the minimization is over a large class of generalized M-estimates of regression with the scale and the intercept parameters being unknown and estimated as nuisance parameters. It also considers the optimal choice of estimates for nuisance parameters.
- Subjects
CHEBYSHEV approximation; HUBER, P. J.; PARAMETER estimation; SYMMETRIC functions; REGRESSION analysis
- Publication
Canadian Journal of Statistics, 1996, Vol 24, Issue 2, p193
- ISSN
0319-5724
- Publication type
Article
- DOI
10.2307/3315625