Psychometrika

, Volume 45, Issue 1, pp 69–97 | Cite as

Principal component analysis of three-mode data by means of alternating least squares algorithms

  • Pieter M. Kroonenberg
  • Jan de Leeuw
Article

Abstract

A new method to estimate the parameters of Tucker's three-mode principal component model is discussed, and the convergence properties of the alternating least squares algorithm to solve the estimation problem are considered. A special case of the general Tucker model, in which the principal component analysis is only performed over two of the three modes is briefly outlined as well. The Miller & Nicely data on the confusion of English consonants are used to illustrate the programs TUCKALS3 and TUCKALS2 which incorporate the algorithms for the two models described.

Key words

three-mode principal component analysis alternating least squares factor analysis multidimensional scaling individual differences scaling simultaneous iteration confusion of consonants 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Reference notes

  1. Carroll, J. D. & Chang, J. J.IDIOSCAL: A generalization of INDSCAL allowing IDIOsyncratic reference systems as well as an analytic approximation to INDSCAL. Paper presented at the Spring Meeting of the Psychometric Society, Princeton, N. J., March 1972.Google Scholar
  2. Harshman, R. A.Foundations of the PARAFAC procedure: Models and conditions for an “explanatory” multimode factor analysis (Working Papers in Phonetics No. 16). Los Angeles: University of California, 1970.Google Scholar
  3. Jennrich, R.A generalization of the multidimensional scaling model of Carroll & Chang (Working Papers in Phonetics No. 22). Los Angeles: University of California, 1972.Google Scholar
  4. Kroonenberg, P. M. & de Leeuw, J.TUCKALS2: A principal component analysis of three mode data (Res. Bull. RB 001-77). Leiden: Department of Data Theory, University of Leiden, 1977.Google Scholar
  5. Levin, J.Three-mode factor analysis (Unpublished doctoral thesis). Urbana, Ill.: University of Illinois, 1963.Google Scholar
  6. Wish, M.An INDSCAL analysis of the Miller & Nicely consonant confusion data. Paper presented at meetings of the Acoustical Society of America. Houston, November, 1970.Google Scholar

References

  1. Carroll, J. D. & Chang, J. J. Analysis of individual differences in multidimensional scaling via an N-way generalization of “Eckart-Young” decomposition.Psychometrika, 1970,35, 283–320.Google Scholar
  2. Carroll, J. D. & Wish, M. Models and methods for three-way multidimensional scaling. In D. H. Krantz, R. D. Luce, R. C. Atkinson, & P. Suppes (Eds.),Contemporary developments in mathematical psychology (Vol. II). San Francisco: W. H. Freeman, 1974.Google Scholar
  3. d'Esopo, D. A. A convex programming procedure.Naval Research Logistics Quarterly, 1959,11, 33–42.Google Scholar
  4. Hubert, L. J. & Baker, F. B. Evaluating the symmetry of a proximity matrix.Quality & Quantity, 1979,13, 77–84.Google Scholar
  5. Israelsson, A. Three-way (or second order) component analysis. In H. Wold & E. Lyttkens (Eds.), Nonlinear iterative partial least-squares (NIPALS) estimation procedures.Bulletin of the International Statistical Institute, 1969,43, 29–51.Google Scholar
  6. Meyer, R. R. The validity of a family of optimization methods.SIAM Journal of Control and Optimization, 1970,15, 699–715.Google Scholar
  7. Miller, G. A. & Nicely, P. E. An analysis of perceptual confusion among some English consonants.Journal of the Acoustical Society of America, 1955,27, 338–352.Google Scholar
  8. Osgood, C. E., Suci, G. J. & Tannenbaum, T. H.The measurement of meaning. Urbana, Ill.: University of Illinois Press, 1957.Google Scholar
  9. Ostrowski, A. M.Solution of equations and systems of equations. New York: Academic Press, 1966.Google Scholar
  10. Penrose, R. On the best approximate solutions of linear matrix equations.Proceedings of the Cambridge Philosophical Society, 1955,51, 406–413.Google Scholar
  11. Rutishauser, H. Computational aspects of F. L. Bauer's simultaneous iteration method.Numerische Mathematik, 1969,13, 4–13.Google Scholar
  12. Sands, R. & Young, F. W. Component models for three-way data: ALSCOMP3, an alternating least squares algorithm with optimal scaling features.Psychometrika, 1980,45, 39–67.Google Scholar
  13. Schwartz, H. R., Rutishauser, H. & Stiefel, E.Numerik. Symmetrischer matrizen. Stuttgart: Teubner, 1968.Google Scholar
  14. Shepard, R. N. Psychological representation of speech sounds. In E. E. David & P. B. Denes (Eds.),Human communication. A unified view. New York: McGraw Hill, 1972.Google Scholar
  15. Shepard, R. N. Representation of structure in similarity data: Problems and prospects.Psychometrika, 1974,39, 373–421.Google Scholar
  16. Smith, P. T. Feature-testing models and their application to perception and memory for speech.Quarterly Journal of Experimental Psychology, 1973,25, 511–534.Google Scholar
  17. Smith, P. T. & Jones, K. F. Some hierarchical scaling methods for confusion matrix analysis II. Application to large matrices.British Journal of Mathematical and Statistical Psychology, 1975,28, 30–45.Google Scholar
  18. Soli, S. D. & Arabie, P. Auditory versus phonetic accounts of observed confusions between consonant phonemes.Journal of the Acoustical Society of America, 1979,66, 46–59.Google Scholar
  19. Takane, Y., Young, F. W. & de Leeuw, J. Non-metric individual differences multidimensional scaling: An alternating least squares method with optimal scaling features.Psychometrika, 1977,42, 7–67.Google Scholar
  20. Tucker, L. R. Implications of factor analysis of three-way matrices for measurement of change. In C. W. Harris (Ed.),Problems in measuring change. Madison, Wis.: University of Wisconsin Press, 1963.Google Scholar
  21. Tucker, L. R. The extension of factor analysis to three-dimensional matrices. In H. Gulliksen & N. Frederiksen (Eds.),Contributions to mathematical psychology. New York: Holt, Rinehardt & Winston, 1964.Google Scholar
  22. Tucker, L. R. Some mathematical notes on three-mode factor analysis.Psychometrika, 1966,31, 279–311.Google Scholar
  23. Tucker, L. R. Relations between multidimensional scaling and three-mode factor analysis.Psychometrika, 1972,37, 3–27.Google Scholar
  24. Tucker, L. R. & Messick, S. An individual difference model for multidimensional scaling.Psychometrika, 1963,28, 333–367.Google Scholar
  25. Wainer, H., Gruvaeus, G. & Blair, M. TREBIG: A 360/75 FORTRAN program for three-mode factor analysis for big data sets.Behavioral Research Methods and Instrumentation, 1974,6, 53–54.Google Scholar
  26. Wainer, H., Gruvaeus, G. & Snijder, F. TREMOD: A 360/75 program for three-mode factor analysis.Behavioral Science, 1971,16, 421–422.Google Scholar
  27. Walsh, J. A. An IBM 709 program for factor analyzing three-mode matrices.Educational and Psychological Measurement, 1964,24, 669–773.Google Scholar
  28. Walsh, J. A. & Walsh, R. A revised Fortran program for three-mode factor analysis.Educational and Psychological Measurement, 1976,36, 169–170.Google Scholar
  29. Young, F. W., de Leeuw, J. & Takane, Y. Quantifying qualitative data. In E. D. Lantermann & H. Feger (Eds.),Similarity and choice. Bern: Huber, 1980 (in press).Google Scholar

Copyright information

© The Psychometric Society 1980

Authors and Affiliations

  • Pieter M. Kroonenberg
    • 1
  • Jan de Leeuw
    • 1
  1. 1.University of LeidenThe Netherlands
  2. 2.Vakgroep W.E.P., Subfakulteit der Pedagogische en Andragogische WetenschappenLeidenThe Netherlands

Personalised recommendations