Pack, P., Jolliffe, Ian T. (1992) Influence in Correspondence-Analysis. Applied Statistics-Journal of the Royal Statistical Society Series C, 41 (2). pp. 365-380. ISSN 0035-9254. (doi:10.2307/2347568) (The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided) (KAR id:22390)
The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided. | |
Official URL: https://doi.org/10.2307/2347568 |
Abstract
The influence of individual observations in a correspondence analysis is investigated. Two alternative definitions of 'observation' are possible. For each definition, expressions are presented for influence on the plotting positions of rows and columns of the contingency table being analysed. Formulae for influence on the corresponding eigenvalues (principal inertias) are also given. There are connections for one definition of observation with results for canonical correlation analysis. An example is discussed in detail.
Item Type: | Article |
---|---|
DOI/Identification number: | 10.2307/2347568 |
Uncontrolled keywords: | Canonical correlation analysis; Contingency tables; Correspondence analysis; Eigenvalues; Eigenvectors; Influence |
Subjects: | Q Science > QA Mathematics (inc Computing science) |
Divisions: | Divisions > Division of Computing, Engineering and Mathematical Sciences > School of Mathematics, Statistics and Actuarial Science |
Depositing User: | M. Nasiriavanaki |
Date Deposited: | 23 Aug 2009 13:59 UTC |
Last Modified: | 05 Nov 2024 10:01 UTC |
Resource URI: | https://kar.kent.ac.uk/id/eprint/22390 (The current URI for this page, for reference purposes) |
- Export to:
- RefWorks
- EPrints3 XML
- BibTeX
- CSV
- Depositors only (login required):