Diagnostics For Principal Components - Influence Functions As Diagnostic-Tools

Brooks, S.P. (1994) Diagnostics For Principal Components - Influence Functions As Diagnostic-Tools. Statistician, 43 (4). pp. 483-494. ISSN 0039-0526. (The full text of this publication is not available from this repository)

The full text of this publication is not available from this repository. (Contact us about this Publication)

Abstract

Care should be taken when interpreting the results of a principal components analysis. As an aid to interpretation and as a guide to reliability, a range of diagnostic tools have been proposed. This paper introduces and reviews recent work on influence functions, as well as introducing new methodology, and highlights their use as diagnostic tools for principal components analysis. A Monte Carlo approach is used to assess significance for the influence values obtained.

Item Type: Article
Subjects: H Social Sciences > HA Statistics
Divisions: Faculties > Science Technology and Medical Studies > School of Mathematics Statistics and Actuarial Science > Statistics
Depositing User: P. Ogbuji
Date Deposited: 04 Jul 2009 07:20
Last Modified: 20 Apr 2012 13:48
Resource URI: http://kar.kent.ac.uk/id/eprint/20404 (The current URI for this page, for reference purposes)
  • Depositors only (login required):