inst/doc/paper/paper.md
In ICS: Tools for Exploring Multivariate Data via ICS/ICA

title: 'ICS: A new implementation and numerically stable algorithms for invariant coordinate selection in R' tags: - R - dimension reduction - exploratory data analysis - numerical stability - scatter matrices date: "24 September 2023" output: html_document: df_print: paged authors: - name: Andreas Alfons orcid: 0000-0002-2513-3788 affiliation: 1 - name: Aurore Archimbaud^[Corresponding author] orcid: 0000-0002-6511-9091 affiliation: 1 - name: Klaus Nordhausen orcid: 0000-0002-3758-8501 affiliation: 2 - name: Anne Ruiz-Gazen orcid: 0000-0001-8970-8061 affiliation: 3 bibliography: paper.bib affiliations: - name: Erasmus School of Economics, Erasmus University Rotterdam, Netherlands index: 1 - name: Department of Mathematics and Statistics, University of Jyväskylä, Finland index: 2 - name: Toulouse School of Economics, Université Toulouse 1 Capitole, France index: 3

Summary

Describe the functionality of the add-on package ICS [@ICS] for the statistical computing environment R [@RCore], similarly to the DESCRIPTION file. Also mention that this paper addresses recent developments while a paper on an older version of the package has already been published [@nordhausen2008].

Statement of need

Outline the new developments: more flexible main function ICS(), numerically more stable algorithms (whitening, QR algorithm), object-oriented handling of scatter matrices. Mention re-implementation using S3 classes instead of S4 classes, as the latter are more accessible to a broader audience of R users. Highlight the relevance of the package by citing papers that have used it.

Related software

Describe related packages such as ICSOutlier [@archimbaud2018] and ICSClust [@ICSClust].

Examples

Illustrate how to use the new functions as well as the advantages of the QR algorithm using the examples from the paper on the QR algorithm [@archimbaud2023].

Acknowledgements

This work was partly supported by a grant of the Dutch Research Council (NWO, research program Vidi, project number VI.Vidi.195.141), by the Austrian Science Fund P31881-N32, by the HiTEc COST Action (CA21163), and by the French Agence Nationale de la Recherche under grant ANR-17-EURE-0010 (Investissements d'Avenir program).