Sankhya: The Indian Journal of Statistics

1997, Volume 59, Series B, Pt. 1, 1-10

CONSISTENCY OF A CLASS OF VARIANCE ESTIMATORS IN LINEAR MODELS UNDER HETEROSCEDASTICITY

By

SHYAMAL D. PEDDADA, University of Virginia, Charlottrsville

And

TODD SMITH, University of Central Arkansas, Conway

 

SUMMARY. For any fixed effects linear model we consider in this article a general variance estimator of an ordinary least squares estimator (OLSE) under heteroscedasticity. Under some conditions on the design matrix, it is shown that the proposed class of estimators consistently estimate the covariance matrix of OLSE.

AMS (1980) subject classification. 62J05

Key words and phrases. Heteroscedasticity, homoscedasticity, linear models, variance estimation.

Full paper (PDF)

This article in Mathematical Reviews