Abstract
Social science shows a growing interest on multivariate models for discrete variables able to capture even complex pattern of relationships among a collection of observable fac-tors. Chain Graph Models use graphs to illustrate and shape these conditional independence assumptions. Furthermore, the different types of association among variables are well represented through directed and undirected arcs. This vi-sual tool is supported from a set of parameters that capture and describe the association. In this work we use these models to study the poverty status and particularly how this one can be affected from a group of selected variables. Several Chain Graph Models on two different cross-section data sets of Hungarian and German Household were tested. The better one for each data set is deeply studied with a par-ticular attention to the parameters denoting the connection between the variables. From this analysis a strong effect of the considered social variables on the poverty status is highlighted in both data-sets.
Original language | English |
---|---|
Publisher | Vita e Pensiero |
Number of pages | 25 |
ISBN (Print) | 978-88-343-3197-2 |
Publication status | Published - 2016 |
Externally published | Yes |
Keywords
- graph models
- income inequality
- marginal log-linear parameters
- marginal models
- welfare system