Un método para determinar observaciones influyentes en la SCE al ajustar modelos de diseños factoriales

This paper presents the generalization of the Qi statistics and the criterion to identify influential observations for the SCE when a linear model Y = Xβ +e of incomplete rank is fitted using factorial design. Throughout the residual analysis, the later statistics is generalised under the classic assu...

Full description

Autores:
Ubaque Lopez, Blanca Cecilia
Rincón Suárez, Luis Francisco
Tipo de recurso:
Fecha de publicación:
2011
Institución:
Universidad Santo Tomás
Repositorio:
Repositorio Institucional USTA
Idioma:
spa
OAI Identifier:
oai:repository.usta.edu.co:11634/39568
Acceso en línea:
https://revistas.usantotomas.edu.co/index.php/estadistica/article/view/58
http://hdl.handle.net/11634/39568
Palabra clave:
Diseño de experimentos
modelo de diseño factorial
suma de cuadrados residual
observaciones influyentes en la SCE
Rights
License
http://purl.org/coar/access_right/c_abf2
Description
Summary:This paper presents the generalization of the Qi statistics and the criterion to identify influential observations for the SCE when a linear model Y = Xβ +e of incomplete rank is fitted using factorial design. Throughout the residual analysis, the later statistics is generalised under the classic assumption ei ∼ N(0;σ2) y Cov(ei,ej) = 0 si i 6= j).