Nordhausen, Klaus and Ruiz-Gazen, Anne (2022) On the usage of joint diagonalization in multivariate statistics. Journal of Multivariate Analysis, vol. 188 (104844).

This is the latest version of this item.

[thumbnail of NordhausenRuizGazen2022.pdf]
Download (1MB) | Preview
Identification Number : 10.1016/j.jmva.2021.104844


Scatter matrices generalize the covariance matrix and are useful in many multivariate data analysis methods, including well-known principal component analysis (PCA), which is based on the diagonalization of the covariance matrix. The simultaneous diagonalization of two or more scatter matrices goes beyond PCA and is used more and more often. In this paper, we offer an overview of many methods that are based on a joint diagonalization. These methods range from the unsupervised context with invariant coordinate selection and blind source separation, which includes independent component analysis, to the supervised context with discriminant analysis and sliced inverse regression. They also encompass methods that handle dependent data such as time series or spatial data.

Item Type: Article
Language: English
Date: March 2022
Refereed: Yes
Uncontrolled Keywords: Blind source separation, Dimension reduction, Independent component analysis, Invariant component selection, Scatter matrices, Supervised dimension reduction
Divisions: TSE-R (Toulouse)
Site: UT1
Date Deposited: 14 Jan 2022 09:28
Last Modified: 20 Nov 2023 14:56
OAI Identifier:

Available Versions of this Item

View Item


Downloads per month over past year