On the Influence Function of the Quintile Share Ratio |
| |
Authors: | Tchilabalo Abozou Kpanzou |
| |
Institution: | 1. Department of Statistics and Actuarial Science, University of Stellenbosch, Matieland, South Africakpanzout@sun.ac.za |
| |
Abstract: | In robust statistics, the influence function was developed as an important measure of sensitivity of estimators to large values. As a measure of income inequality, the quintile share ratio was introduced and not much is known about the theoretical properties of its nonparametric estimator. One such property is its sensitivity to outliers. In this article, we derive the influence function of the quintile share ratio. As is to be expected from its definition, the influence function is unbounded. A nonparametric estimator for the quintile share ratio is defined and its sensitivity to outliers is investigated in a small simulation study. |
| |
Keywords: | Asymptotic variance Income inequality Nonparametric estimator Robust statistics Sensitivity |
|
|