We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
WELFARE RANKINGS IN THE PRESENCE OF CONTAMINATED DATA.
- Authors
Cowell, Frank A.; Victoria-Feser, Maria-Pia
- Abstract
The article focuses welfare rankings in the presence of contaminated data. Ranking theorems based on concepts of stochastic dominance are fundamental to the analysis of income distributions as well as to finance theory and other aspects of the theory of choice under uncertainty. As abstract theoretical constructs they provide a connection between the philosophical basis of welfare judgments and elementary statistical tools for describing distributions. In practical applications they suggest useful ways in which simple computational procedures may be used to draw inferences from collections of empirical income distributions. However, formal welfare propositions can only be satisfactorily invoked for empirical constructs if sample data can be taken as a reasonable representation of the underlying income distributions that we want to compare. The article specifies explicit objective function such as a utility function or social-welfare function: this has the attraction of simplicity but is rather restrictive. The second approach involves specifying a class of measures of dispersion, and using information about dispersion and mean income jointly to draw conclusions about welfare: but this runs into a class of problems that arise in connection with the estimation of inequality measures and other measures of dispersion.
- Subjects
INCOME inequality; WELFARE economics; DISTRIBUTION (Probability theory); RANKING (Statistics); DISPOSABLE income; INCOME redistribution
- Publication
Econometrica, 2002, Vol 70, Issue 3, p1221
- ISSN
0012-9682
- Publication type
Article
- DOI
10.1111/1468-0262.00324