We present the asymptotic properties of double-stage quantile regressionestimators with random regressors, where the first stage is based on quantile regressionswith the same quantile as in the second stage, which ensures robustness of the estimationprocedure. We derive invariance properties...

...In this paper, we tackle all these difficulties by conducting a full geomatrical analysis of the comparative statics of general models...

Using parametric formulae under lognormality for a broad family of povertymeasures, we show that when inequality measured by the Gini coeffcient isconstant, dening the poverty line as a fraction of a central tendency of theliving standard distribution restricts the evolution of the poverty...

We present in this paper the asymptotic properties of two-stage quantile regressionestimators. These results permit valid inferences in structural models estimated using quantileregressions, in which the possible endogeneity of some explanatory variables is treated viaancilliary predictive...

A priori knowledge about the shape of living standards distribution has notbeen fully exploited in the literature to investigate properties of poverty indices.The method we propose is to exploit credible distributional assumptions to: generateadditional properties for poverty indices; relate the...

We derive an explicit formula of the Watts’ poverty index, in terms of parametersof bivariate lognormal distributions of price indices and nominal livingstandards. This result enables us to: analyse the contributions of the distributionsof prices and nominal living standards in poverty;...