A Basic Result on the Consistency of Maximum Likelihood Estimators

A sequence fXig of independent and identically distributed random objects is considered. The common distribution of the X i's is absolutely continuous with respect to a givenmeasure, and the corresponding density is not completely speciØed but depends on an unknown parameter. Under mild topolog...

Descripción completa

Detalles Bibliográficos
Autor: GRACIELA MARIA DE LOS DOLORES GONZALEZ FARIAS
Tipo de recurso: informe técnico
Estado:Versión publicada
Fecha de publicación:2007
País:México
Institución:Centro de Investigación en Matemáticas
Repositorio:Repositorio Institucional CIMAT
Idioma:inglés
OAI Identifier:oai:cimat.repositorioinstitucional.mx:1008/637
Acceso en línea:http://cimat.repositorioinstitucional.mx/jspui/handle/1008/637
Access Level:acceso abierto
Palabra clave:info:eu-repo/classification/MSC/Estimación
info:eu-repo/classification/cti/1
info:eu-repo/classification/cti/12
info:eu-repo/classification/cti/1209
info:eu-repo/classification/cti/120909
Descripción
Sumario:A sequence fXig of independent and identically distributed random objects is considered. The common distribution of the X i's is absolutely continuous with respect to a givenmeasure, and the corresponding density is not completely speciØed but depends on an unknown parameter. Under mild topological and continuity requirements, a necessary and su±cient crite- rion for the consistency of a sequence of maximum likelihood estimators is obtained. When this characterization is applied to the case in which the parameter belongs to a Ønite dimensional Euclidean space, the conclusion is that the sequence is consistent if and only if it is bounded with probability .1.