Partition entropy and chi-squared error: the improved MEMPHIS algorithm - Part I

The entropy of the population partition is studied as a function of the sampling parameter, so that within a particular interval of its graph, the plateau region, it is possible to get a stable estimation of the mixture parameters. The optimal estimation is associated with a local maximum of entropy...

Descripción completa

Detalles Bibliográficos
Autores: Cubarsí Morera, Rafael|||0000-0001-7748-1322, Alcobé López, Santiago
Tipo de recurso: informe técnico
Fecha de publicación:2009
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/6775
Acceso en línea:https://hdl.handle.net/2117/6775
Access Level:acceso abierto
Palabra clave:Astronomy and astrophysics
Probabilities
Stochastic processes
Mathematical statistics
Astronomia
Astrofísica
Probabilitats
Processos estocàstics
Estadística matemàtica
85 ASTRONOMY AND ASTROPHYSICS
60 PROBABILITY THEORY AND STOCHASTIC PROCESSES
62 STATISTICS
Àrees temàtiques de la UPC::Matemàtiques i estadística::Estadística matemàtica::Mètodes estadístics
Descripción
Sumario:The entropy of the population partition is studied as a function of the sampling parameter, so that within a particular interval of its graph, the plateau region, it is possible to get a stable estimation of the mixture parameters. The optimal estimation is associated with a local maximum of entropy. Alter natively, the $\chi^2$ error of the mixture approach may also be used to obtain an optimal segregation. The relationship between the fitting error and the population entropy has been analysed in detail. We have proved that, by using an appropriate sampling parameter, within a plateau region of the entropy graph, a local entropy maximum takes place simultaneously with a local minimum of the $\chi^2$ error. Therefore, the combined statistical method provides the best approximation mixture, as well as the less informative partiti on, to estimate the kinematic parameters of populations.