Premium
Mettre au test la dominance des états de pauvreté: une application au Canada .
Author(s) -
Chen WenHao,
Duclos JeanYves
Publication year - 2011
Publication title -
canadian journal of economics/revue canadienne d'économique
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.773
H-Index - 69
eISSN - 1540-5982
pISSN - 0008-4085
DOI - 10.1111/j.1540-5982.2011.01654.x
Subject(s) - poverty , stochastic dominance , econometrics , dominance (genetics) , ranking (information retrieval) , statistics , sampling (signal processing) , range (aeronautics) , normative , statistical hypothesis testing , economics , mathematics , computer science , political science , economic growth , engineering , artificial intelligence , biochemistry , chemistry , filter (signal processing) , aerospace engineering , law , computer vision , gene
The paper proposes and applies statistical tests for poverty dominance that check for whether poverty comparisons can be made robustly over ranges of poverty lines and classes of poverty indices. This helps provide both normative and statistical confidence in establishing poverty rankings across distributions. The tests, which can take into account the complex sampling procedures that are typically used by statistical agencies to generate household‐level surveys, are implemented using the Canadian Survey of Labour and Income Dynamics (SLID) for 1996, 1999, and 2002. Although the yearly cumulative distribution functions cross at the lower tails of the distributions, the more recent years tend to dominate earlier years for a relatively wide range of poverty lines. Failing to take into account SLID’s sampling variability (as is sometimes done) can inflate significantly one’s confidence in ranking poverty. Taking into account SLID’s complex sampling design (as has not been done before) can also decrease substantially the range of poverty lines over which a poverty ranking can be inferred.