No Detail Found!

Impacts of error variance on an asymptotically unbiased estimator for inverse regression.

Author(s)
M. Y. Ashkar Department of Mathematics, University of Bahrain P.O. BOX 32038, Bahrain
M. A. Ali Department of Mathematics, University of Bahrain P.O. BOX 32038, Bahrain
Abstract
Srivastava and Singh (1989) proposed a class of weighted average estimators for inverse linear regression. They examined the asymptotic bias and the asymptotic mean squared error (MSE) of this family of estimators. It was found that a member of this class is asymptotically unbiased to the order of their approximation. This article compares the well known classical, inverse and the asymptotically unbiased weighted average estimators of inverse linear regression by Monte Carlo methods. It is observed that for small and moderate size samples the asymptotically unbiased weighted average estimator is relatively interior to the inverse estimator from a mean squared error point of view.
Publication Details
Page(s) 263-275
DOI DOI not available
Published Journal: Pakistan Journal of Statistics, Volume: 20, Issue: 2, Year: 2004

Request the full text from PASTIC

0

Citations

0

Downloads

38

Views