Component-wise dimension reduction
| Autor: | Valerii V. Fedorov, Agnes M. Herzberg, Sergei L. Leonov |
|---|---|
| Rok vydání: | 2003 |
| Předmět: |
Statistics and Probability
Variables Covariance matrix Applied Mathematics Dimensionality reduction media_common.quotation_subject Design matrix Exploratory factor analysis Dimension (vector space) Statistics Principal component analysis Statistics Probability and Uncertainty media_common Mathematics Factor analysis |
| Zdroj: | Journal of Statistical Planning and Inference. 114:81-93 |
| ISSN: | 0378-3758 |
| DOI: | 10.1016/s0378-3758(02)00464-0 |
| Popis: | Principal components methods and factor analysis are popular tools for the dimension-reduction problem. These techniques can be used to obtain a smaller number of new variables. However, the new variables may include all or most of the original variables. In this study, two methods are given which will select the most informative subset of variables from the variables which are directly measured. The different approaches are compared in a concluding example. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect