You are not currently logged in.
Access JSTOR through your library or other institution:
On the Bounded-Influence Regression Estimator of Krasker and Welsch
Journal of the American Statistical Association
Vol. 80, No. 389 (Mar., 1985), pp. 205-208
Stable URL: http://www.jstor.org/stable/2288073
Page Count: 4
You can always find the topics here!Topics: Estimators, Statistical estimation, Statistical variance, Covariance, Least squares, Analytical estimating, Coordinate systems, Maximum likelihood estimation, Design efficiency, Linear regression
Were these topics helpful?See somethings inaccurate? Let us know!
Select the topics that are inaccurate.
Preview not available
Recently, Krasker and Welsch (1982) considered a class of bounded-influence regression estimators. They showed that within this class the so-called Krasker-Welsch estimator is the only solution to a first-order necessary condition for strong optimality, that is, for minimizing, in the sense of positive definiteness, the asymptotic covariance matrix. However, whether any strongly optimal estimator in fact exists remained an open question. In this article, an example is given where no strongly optimal estimator exists. Moreover, the practical significance of the lack of a strongly optimal estimator is discussed.
Journal of the American Statistical Association © 1985 American Statistical Association