Estrategia para la selección de modelos gráficos en tablas de contingencia no estructuradas

Loglinear models are frequently utilized in the analysis of association patterns betweenvariables in a multidimensional contingency table. Discrete graphical modelsare fashionable. They are a subset of the hierarchical loglinear models and its conditionalindependence structures are represented by a...

Descripción completa

Autores Principales: González Debén, Adalberto, Alfonso Sagué, Karen
Formato: Artículo
Idioma: Español
Publicado: 2015
Acceso en línea: http://revistas.ucr.ac.cr/index.php/matematica/article/view/223
http://hdl.handle.net/10669/12866
Sumario: Loglinear models are frequently utilized in the analysis of association patterns betweenvariables in a multidimensional contingency table. Discrete graphical modelsare fashionable. They are a subset of the hierarchical loglinear models and its conditionalindependence structures are represented by a graph. Because of that they canbe more easily interpreted. In this work four models selection methods are describedand compared: stepwise selection (backward, forward and in two steps) and the EHprocedure.A strategy for model selection is proposed.Keywords: graphical models, contingency tables, model selection methods