Most recent version is MMX2ER16, which has been resubmitted to Richard Smith, Econometrics Journal on 7th April 2012

A Minimax Bias Heteroskedasticity Consistent Variance Estimate for OLS

Mumtaz Ahmed[1] & Asad Zaman[2]

[1]Mumtaz Ahmed is a PhD Student in IIIE, International Islamic University Islamabad, Pakistan.

Email: mumtaz_phd@yahoo.com

[2]Asad Zaman is a professor of Economics in IIIE, International Islamic University Islamabad, Pakistan.

Email: asadzaman@alum.mit.edu

Abstract

Analytic evaluation of heteroskedasticity consistent covariance matrix estimates (HCCME) is difficult because of the complexity of the formulae currently available. We obtain analytic formulae for the bias of some HCCME’s in the simplest case of a single regressor. These provide substantial insight into the properties and performance characteristics of these estimators. In particular, we provide a new estimator which minimizes the maximum possible bias and improves substantially on the standard Eicker-White estimate.