Heteroskedasticity and Neglected Parameter Heterogeneity
| Published date | 01 May 2001 |
| DOI | http://doi.org/10.1111/1468-0084.00220 |
| Author | Joachim Zietz |
| Date | 01 May 2001 |
practitioners corner
Heteroskedasticity and neglected parameter
heterogeneity
Joachim Zietzy
Professor of Economics, Department of Economics and Finance, Middle Tennessee
State University
I. Introduction
Evidence of heteroskedasticity is often treated with little concern by applied
researchers. This is not surprising. According to the standard textbook
regression model, heteroskedasticity will not affect the estimated regression
coef®cients but only their standard errors. In addition, convenient corrections
are readily available for heteroskedasticity, such as White's (1980) hetero-
skedasticity ± consistent variance covariance matrix estimator (HCVE). The
purpose of this note is to introduce to a broader audience that the routine
application of White's HCVEs can be a serious mistake, because it may result
in meaningless estimates of both the regression coef®cients and their
standard errors.1As demonstrated in this paper, this conclusion results from
the fact that heteroskedasticity may arise from neglected parameter hetero-
geneity in the standard linear regression model.
The remainder of this paper is organized as follows. The next section
introduces the concept of neglected parameter heterogeneity, its potential
statistical consequences, and some tests that have been suggested in the
literature. The following section summarizes the set-up and results of a series
of Monte-Carlo experiments that illustrate more concretely how neglected
parameter heterogeneity impacts regression coef®cients and test statistics for
heteroskedasticity and for some other popular misspeci®cation tests. The
paper concludes with a brief summary of its main points.
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 63, 2 (2001) 0305-9049
#Blackwell Publishers Ltd, 2001. Published byBlackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UKand 350
Main Street, Malden, MA 02148, USA.
263
yThe author wishes to thank an anonymous referee and, in particular, Jonathan Temple for helpful
comments.
1The point as such is not new.See, for example, Temple (1999, fn. 17).
To continue reading
Request your trial