A simulation study on nonlinear principal component analysis

Brigitta Voß

No 1999,42, Technical Reports from Technische Universität Dortmund, Sonderforschungsbereich 475: Komplexitätsreduktion in multivariaten Datenstrukturen

Abstract: In statistical practice multicollinearity of predictor variables is rather the rule than the exception and appropriate models are needed to avoid instability of predictions. Feature extraction methods reflect the idea that latent variables not measurable directly are underlying the original data. They try to reduce the dimension of the data by constructing new independent variables which keep as much information as possible from the original measurements. A common feature extraction method is Principal Component Analysis (PCA), which in its classical form is restricted to linear relationships among predictor variables. This paper is concerned with nonlinear principal component analysis (NLPCA) as introduced by Kramer (1991) who modelled his approach with help of artificial neural networks. By means of first simulation studies data derived from semicircles and circles are investigated with respect to their ability to be described by nonlinear principal components among the predictors.

Keywords: feature extraction; nonlinear principal component analysis; artificial neural networks (search for similar items in EconPapers)
Date: 1999
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.econstor.eu/bitstream/10419/77363/2/1999-42.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:zbw:sfb475:199942

Access Statistics for this paper

More papers in Technical Reports from Technische Universität Dortmund, Sonderforschungsbereich 475: Komplexitätsreduktion in multivariaten Datenstrukturen Contact information at EDIRC.
Bibliographic data for series maintained by ZBW - Leibniz Information Centre for Economics ().