Robust Multidimensional Poverty Comparisons
We investigate how to make poverty comparisons using multidimensional indicators of well-being, showing in particular how to check whether the comparisons are robust to the choice of poverty indices and poverty lines. Our methodology applies equally well to either of what can be defined as "union" and "intersection" approaches to dealing with multidimensional indicators of well-being. When one of two variables is discrete, our methods specialize to those that Atkinson (1991), Jenkins and Lambert (1993) and others have developed to deal with household composition heterogeneity. The results also extend the statistical results recently derived in Davidson and Duclos (2000) to cases where well-being is measured in two or more dimensions. We thus derive the sampling distribution of various multidimensional poverty estimators, including estimators of the "critical" frontiers of poverty lines above which multidimensional poverty comparisons are no longer ethically robust.
|Date of creation:||2001|
|Date of revision:|
|Contact details of provider:|| Postal: Pavillon J.A. De Sève, Québec, Québec, G1K 7P4|
Phone: (418) 656-5122
Fax: (418) 656-2707
Web page: http://www.ecn.ulaval.ca
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:lvl:laeccr:0115. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Manuel Paradis)
If references are entirely missing, you can add them using this form.