Model based approach for household clustering with mixed scale variables

摘要: The Ministry of Social Development in Mexico is charge creating and assigning social programmes targeting specific needs the population for improvement quality life. To better target programmes, aimed to find clusters households with same based on demographic characteristics as well poverty conditions household. Available data consists continuous, ordinal, nominal variables observations are not iid but come from a survey sample complex design. We propose Bayesian nonparametric mixture model that jointly models this mixed scale accommodates different sampling probabilities. performance assessed via simulated data. A full analysis socio-economic State presented.