首页 | 本学科首页   官方微博 | 高级检索  
   检索      


Influential observations in principal factor analysis
Authors:Yutaka Tanaka  Yoshimasa Odaka
Institution:(1) Department of Statistics, Okayama University, 2-1-1 Tsushima-naka, 700 Okayama, Japan;(2) Information Processing Center, Okayama University of Science, Japan
Abstract:We propose a method for detecting influential observations in iterative principal factor analysis. For this purpose we derive the influence functionsI(x; LL T ) andI(x; delta) for the common variance matrixT =LL T and the unique variance matrix Delta, respectively, in the common factor decomposition Sgr =LL T + Delta. A numerical example is given for illustration.The authors are grateful to Tomoyuki Tarumi and Atsuhiro Hayashi for their kind permission to use their software Seto/B for drawing Figures 1 and 2 and to anonymous reviewers for comments on the paper.
Keywords:influential observations  influence function  influence on the subspace spanned by eigenvectors  principal factor analysis  perturbation theory of eigenvalue problems
本文献已被 SpringerLink 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号