Bidimensional Asymptotic Normality of the Adaptive Moving Kernel Poverty Index Estimate

Abstract

In this paper, we study the adaptive kernel estimator for the bidimensional extension of Foster, Greer and Thorbecke class of measures. The asymptotic normality of the estimator is established. Next, we show how the proposed estimator can generate sequential confidence intervals by a moving adaptive kernel process. As an illustration, we determine the confidence intervals for different regions of Senegal. The study of this application demonstrated that our methodology is not only more efficient than the classical and empirical estimator, but it also provides better confidence intervals for the poverty index