A Short Comment on the use of R_adj^2 in Social Science

It is a common practice to prefer , over  to assess the explainability power of a statistical regression model among social scientists, especially for one having more than one independent variables. However, this preference is not advantageous at all times because the usage of  may end up in negativ...

Descripción completa

Detalles Bibliográficos
Autor: Siddiqi, Ahmed F.
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2019
País:Ecuador
Institución:Universidad San Gregorio de Portoviejo
Repositorio:Revista Universidad San Gregorio de Portoviejo
Idioma:inglés
OAI Identifier:oai:ojs.revista.sangregorio.edu.ec:article/890
Acceso en línea:http://201.159.222.49/index.php/REVISTASANGREGORIO/article/view/890
Access Level:acceso abierto
Palabra clave:r^2; r_adj^2
Descripción
Sumario:It is a common practice to prefer , over  to assess the explainability power of a statistical regression model among social scientists, especially for one having more than one independent variables. However, this preference is not advantageous at all times because the usage of  may end up in negative coefficients making them non-interpretable. A Monte Carlo simulation experiment is used to appraise the behavior of these adjusted versions of for different numbers of independent variables. It has been found that almost all of the selected adjusted version of  produces negative coefficients.

Similares en LA Referencia