K. Zografos, G. Avlogiaris, A. Micheas
Divergence measures formulate the amount of information contained in the data for discrimination between two probabilistic models. They obey some of the properties of a distance between two probability distributions and, in this sense, they have been extensively applied in several disciplines and contexts to develop procedures for estimation and testing of hypotheses. The aim of this work is to introduce measures of divergence in a local setting, that is, to propose pseudo-distances between two distributions in a specific area of their joint domain. The proposed measures will be applied for testing statistical hypotheses, in a local framework.
Palabras clave: φ-divergence, Cressie and Read power Divergence, Local divergence, Exponential family, Local φ -divergence test, Local homogeneity test
Programado
X05.3 Inferencia Estadística I
7 de septiembre de 2016 12:30
Aula 21.08
