The paper deals with statistical validation of the Millennium Development Goals, especially the one targeting halving the extreme poverty in 2015. That we are able, given poverty measure J, to estimate its exact variation from one period s to another t. Clearly, this should be based on longitudinal data.

Lo et al. [1] proposed a general form of the poverty measures, named the General Poverty Index (GPI), including the available ones and conducted in [2] a general asymptotic theory for time-dependents incomes variables. Based on these results, we propose methods of estimation the poverty variation and poverty bridging, simulate the methods and conduct data-driven applications to West African data.

The simulation studies include copula modeling with Singh-Maddala margins and several classical copulas are used. The scripts of the applications are provided and adaptable to any database.

References:

[1] Gane Samb Lô, Serigne Touba Sall et Cheikh Tidiane Seck. Une théorie Générale Asymptotique des Mesures de Pauvreté, (2009), C. R. Math. Rep. Acad. Sci. Canada, 45-52, Volume: 31 (2),

[2] Gane Samb Lô and Serigne Touba Sall. Asymptotic representation theorems for poverty indices, Afrika Statistika, Volume 5, pp. 238-244

Keywords: Poverty measures; income distribution; empirical process; time-dependent asymptotic normality

Biography: Mr Mergane is young statistician from Saint-Louis University preparing Ph.D dissertation.