A Monte Carlo Study of the Accuracy and Robustness of Ten Bivariate Location Estimators

by Jean-Claude Masse and Jean-Francois Plante .

Abstract: In a Monte Carlo study, we compare ten bivariate location estimators as regards their closeness to the location parameter, accuracy and robustness. The behavior of the estimators is examined under various sampling situations determined by four sample sizes and fourteen underlying distributions. In addition to the arithmetic mean, five bivariate medians and four depth-based trimmed means are investigated. The performance of each estimator is assessed through numerical functions of the sample mean squared error and bias matrices.

Key Words: Bivariate Location Estimators, Bivariate Medians, Bivariate Trimmed Means, Robustness

Jean-Claude Masse, jcmasse@mat.ulaval.ca
Jean-Francois Plante, jplante@mat.ulaval.ca

Editor: Joseph W. McKean , joe@stat.wmich.edu

READING THE ARTICLE: You can read the article in portable document (.pdf) format (548741 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 2401 times since July 24, 2006.

Return to the InterStat Home Page.