Bootstrapping inequality measures under the null hypothesis: Is it worth the effort?

This paper discusses methods of statistical inference for inequality measures, in particular the nonparametric bootstrap. Standard resampling techniques and a new method for nonparametric resampling under the null hypothesis are discussed. Monte-Carlo simulations show that some bootstrap methods outperform the commonly used normal approximation while other bootstrap methods—including those which are used in most empirical applications—are hardly any better. Copyright Springer-Verlag 2002