Procrustes analysis applied to variables selection
In exploratory multivariate research aiming at the reduction of the, dimension of the variables set, the most frequently used method is the analysis of the principal components. All original variables are generally necessary to define the subset of variables. Krzanowski (1987) has provided a methodo...
| Autores: | , |
|---|---|
| Formato: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2008 |
| País: | Brasil |
| Recursos: | Universidade Estadual de Maringá (UEM) |
| Repositorio: | Acta scientiarum. Technology (Online) |
| Idioma: | portugués |
| OAI Identifier: | oai:periodicos.uem.br/ojs:article/3073 |
| Acesso em linha: | http://www.periodicos.uem.br/ojs/index.php/ActaSciTechnol/article/view/3073 |
| Access Level: | acceso abierto |
| Palavra-chave: | componentes principais análise procrustes análise multivariada 1.02.02.00-5 Estatística |
| Resumo: | In exploratory multivariate research aiming at the reduction of the, dimension of the variables set, the most frequently used method is the analysis of the principal components. All original variables are generally necessary to define the subset of variables. Krzanowski (1987) has provided a methodology which combines the principal component analysis and the procrustes analysis to determine how much the new subset of variables reproduces the structure of original variables. Steiner (1995) used several methods to separate groups and select the variables in a medical case study. In the present work, the procrustes analysis was applied to a set of data randomly generated according to the variables distributions defined by Steiner. The objective was to verify if the subset of variables resultant from the analysis reproduces the original structure of data. The: results led to the conclusion that the procrustes method is a necessary tool for variables selection in multivariate analysis. |
|---|