Note on heteroscedasticity in cross-section regressions estimated from Irish county data

Abstract

It is well knqwn that, even when they are expressed in log or per capita form, many statistical series on Irish counties (by which is meant here the twenty-six counties of the Republic) are characterised by a value for Dublin which is considerably greater than the values for all other counties. The fact that the Dublin observation is an 'outlier' does not of itself pose a problem for an econometrician who wishes to estimate cross-section regressions from Irish county data. However, a problem does arise when the inclusion of Dublin makes a significant difference to estimated parameters. The purpose of the present note is to show that, in at least one case where this common phenomenon occurs, the frequently adopted procedure of accepting the equation which excludes Dublin may be given a statistical justification. Specifically, parameter estimates obtained in this way are shown to be statistically indistinguishable from those estimated by applying a heteroscedastic correction procedure to the equation estimated from observations on all twenty-six counties.