Principal component analysis in an asymmetric norm (Q2418501)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Principal component analysis in an asymmetric norm |
scientific article |
Statements
Principal component analysis in an asymmetric norm (English)
0 references
27 May 2019
0 references
The generalization of the PCA is proposed for quantiles and expectiles. Under the elliptically symmetric distributions it turns to the convential PCA. The proposed approach is suitable to study extremes of the multivariate data. Experiments with weather datasets illustrate the efficiency of the proposed method.
0 references
dimension reduction
0 references
quantile
0 references
asymmetric norm
0 references
expectile
0 references
risk attitude
0 references
growth data
0 references
0 references