Premium
Lorenz Curve Inference with Sample Weights: An Application to the Distribution of Unemployment Experience
Author(s) -
Beach C. M.,
Kaliski S. F.
Publication year - 1986
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.2307/2347862
Subject(s) - lorenz curve , inference , sample (material) , econometrics , mathematics , unemployment , statistics , distribution (mathematics) , economics , mathematical economics , computer science , mathematical analysis , gini coefficient , inequality , economic inequality , chromatography , artificial intelligence , chemistry , economic growth
SUMMARY This paper extends the recent methodology of Beach and Davidson, 1983, on performing statistical inference with Lorenz curves and quantile shares to samples which involve weighted observations. Examples of such survey samples are the Current Population Survey in the United States, the Survey of Personal Incomes in the U.K. and the Labour Force Survey in Canada. The paper presents an efficient computational algorithm and provides appropriate inference formulas (equations (1 )‐(3)) which are directly applicable to standard national surveys. An example is presented that involves a novel application of Lorenz curves to the distribution of unemployment experience. This allows one to analyze in disaggregative fashion the distribution of the burden of unemployment across workers in the economy. The distribution of unemployment is shown to be highly unequal and to differ significantly between men and women, particularly among those experiencing the least and greatest amounts of unemployment.