Summary
This paper proposes the generalized lambda distribution (GLD) as a model for describing the distribution of income over a population. The performance of various methods of fitting the GLD to grouped income data are evaluated. Of the estimators considered it is concluded that the unweighted least squares regression on group means should be used.

keywords
distribution fitting, grouped data, inequality indices

COMPSTAT 2004 - Proceedings in Computational Statistics
16th Symposium Held in Prague, Czech Republic, 2004
Antoch Jaromir (Ed.). Physica-Verlag (2004), p.1861-1867.

Back to main publications	Home page	Adobe pdf