Calculating a Standard Error for the Gini Coefficient: Some Further Results

Various authors have proposed using the jackknife technique to approximate a standard error for the Gini coefficient. It has also been shown that the Gini measure can be obtained simply from an artificial OLS regression based on the data and their ranks. Accordingly, we show that obtaining an exact analytical expression for the standard error is a trivial matter. In addition, by extending the regression framework to one involving Seemingly Unrelated Regressions, several interesting hypotheses regarding the sensitivity of the Gini coefficient to changes in the data are readily tested in a formal manner.