Building generalized linear models for ordinal data
Editorial: Universidad de Murcia. Departamento de Estadística e Investigación Operativa
ISBN: 978-84-691-8159-1
Año de publicación: 2009
Congreso: Congreso Nacional de Estadística e Investigación Operativa (31. 2009. Murcia)
Tipo: Aportación congreso
Resumen
In addition to tting the whole model for the specic type of analysis, dierent methods for automatic model building can be employed in analysis using the generalized linear model. Specically, forward entry, backward removal, forward stepwise, and backward stepwise procedures can be performed, as well as best-subset search procedures. Stepwise selection in generalized linear regression based on Wald and score tests is described in Fahrmeir and Frost (1992). In this work, on one hand, a new selection criterion based on divergence measures is proposed in backward removal variable selection for an ordinal generalized linear model. And, on the other hand, a new two-phase model building procedure is built. An application to real data set is given.