The assumption of proportional components when Candecomp is applied to symmetric matrices in the context of INDSCAL

Mohammed Bennani Dosse*, Jos M. F. Ten Berge

*Corresponding author for this work

    Research output: Contribution to journalArticleAcademicpeer-review

    7 Citations (Scopus)

    Abstract

    The use of Candecomp to fit scalar products in the context of INDSCAL is based on the assumption that the symmetry of the data matrices involved causes the component matrices to be equal when Candecomp converges. Ten Berge and Kiers gave examples where this assumption is violated for Gramian data matrices. These examples are believed to be local minima. It is now shown that, in the single-component case, the assumption can only be violated at saddle points. Chances of Candecomp, converging to a saddle point are small but still nonzero.

    Original languageEnglish
    Pages (from-to)303-307
    Number of pages5
    JournalPsychometrika
    Volume73
    Issue number2
    DOIs
    Publication statusPublished - Jun-2008

    Keywords

    • Candecomp
    • Parafac
    • Indscal

    Cite this