Un método para determinar un grupo de observaciones influyentes en la SCE al ajustar modelos de rango completo
Throughout this paper is presented a method that allows us to identify the presence of an influential group of observations for the RSS after fitting a linear model of full rank. Over the residual analysis a Qi statistic is used and the criteria is built, using the distribution of the statistic mentio...
- Autores:
-
Rincón Suárez, Luis Fransisco
- Tipo de recurso:
- Fecha de publicación:
- 2010
- Institución:
- Universidad Santo Tomás
- Repositorio:
- Repositorio Institucional USTA
- Idioma:
- spa
- OAI Identifier:
- oai:repository.usta.edu.co:11634/39535
- Acceso en línea:
- https://revistas.usantotomas.edu.co/index.php/estadistica/article/view/25
http://hdl.handle.net/11634/39535
- Palabra clave:
- Modelo de rango completo
observaciones influyentes
parámetros estimados
suma de cuadrados residual
- Rights
- License
- http://purl.org/coar/access_right/c_abf2
| Summary: | Throughout this paper is presented a method that allows us to identify the presence of an influential group of observations for the RSS after fitting a linear model of full rank. Over the residual analysis a Qi statistic is used and the criteria is built, using the distribution of the statistic mentioned under the assumption than ei ∼ N(0;σ2) and Cov(ei,ej) = 0 if (i 6= j). |
|---|