Variance estimation for sensitivity analysis of poverty and inequality measures

  • Christian Dudel Ruhr-University Bochum, Max Planck Institute for Demographic Research
Keywords: equivalence scale, influence function, low income proportion, sensitivity analysis, variance estimation,


Estimates of poverty and inequality are often based on application of a single equivalence scale, despite the fact that a large number of different equivalence scales can be found in the literature. This paper describes a framework for sensitivity analysis which can be used to account for the variability of equivalence scales and allows to derive variance estimates of results of sensitivity analysis. Simulations show that this method yields reliable estimates. An empirical application reveals that accounting for both variability of equivalence scales and sampling variance leads to confidence intervals which are wide.