Parallel Factor Analysis with Constraints on the Configurations: An overview

  • Pieter M. Kroonenberg
  • Willem J. Heiser
Conference paper
Part of the Studies in Classification, Data Analysis, and Knowledge Organization book series (STUDIES CLASS)


The purpose of the paper is to present an overview of recent developments with respect to the use of constraints in conjunction with the Parallel Factor Analysis PARAFAC model (Harshman, 1970). Constraints and the way they can be incorporated in the estimation process of the model are reviewed. Emphasis is placed on the relatively new triadic algorithm which provides a large number of new ways to use the PARAFAC model.


Discrepancy Function Positive Matrix Factorization Nonnegativity Constraint Orthogonality Constraint PARAFAC Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Carroll, J.D. (1987): New algorithm for symmetric cANDECOMP. Unpublished manuscript. ATandT Bell Laboratories, Murray Hill, NJ.Google Scholar
  2. Carroll, J.D. and Chaturvedi, A. (1995): A general approach to clustering and multidimensional scaling of two-way, three-way, and higher way data. Geometric representations of perceptual phenomena: Papers in honor of Tarow Indow on his 70th birthday, Luce, R.D. et al. (eds.), Erlbaum. -Mahwah, NJ.Google Scholar
  3. Carroll; J.D. and Chang, J.-J. (1970): Analysis of individual differences in multidimensional scaling via an N-way generalization of “Eckart-Young” decomposition. Psychometrika,35. 283–319.Google Scholar
  4. Carroll, J.D., De Soete, C., and Kamenski, A.D. (1992): A modified cANDECOMP algorithm for fitting the latent class model: Implementation and evaluation. Applied Stochastic Models and Data Analysis, 8, 303–309.CrossRefGoogle Scholar
  5. Carroll, J.D., Pruzansky, S., and Kruskal, J.B. (1980): CANDELINC: A general approach to multidimensional analysis of many-way arrays with linear constraints on parameters. Psychometrika, 45, 3–24.MathSciNetMATHCrossRefGoogle Scholar
  6. Denis, J.B. and Dhorne, T. (1989): Orthogonal tensor decomposition of 3-way tables. In: Multiway data analysis, Coppi, R. and Bolasco, S. (eds.), 31–38, Elsevier, Amsterdam.Google Scholar
  7. DeSarbo, W.S., Carroll, J.D., Lehmann, D.R., and O’Shaughnessy, J. (1982). Three-way multivariate conjoint analysis. Marketing Science, 1, 323–350.CrossRefGoogle Scholar
  8. Durrell, S.R., Lee, C.-H., Ross, R.T., and Gross, E.L. (1990): Factor analysis of the near-ultraviolet absorption spectrum of plastocyanin using bilinear, trilinear, and quadrilinear models. Archives of Biochemistry and Biophysics, 278, 148–160.CrossRefGoogle Scholar
  9. Franc, A. (1992): Etude algébrique des multitableaux: Apports de l’algèbre tensorielle. Unpublished PhD thesis, Université de Montpellier II, France.Google Scholar
  10. Gifi, A. (1990): Nonlinear multivariate analysis, Wiley, Chicester, UK.Google Scholar
  11. Greenacre, M.J. Theory and applications of correspondence analysis,Academic Press, London.Google Scholar
  12. Harshman, R.A. (1970): Foundations of the PARAFAC procedure: Models and contributions for an “explanatory” multi-modal factor analysis. UCLA Working Papers in Phonetics, 16, 1–84. [Also available as University Microfilms, No. 10,0085].Google Scholar
  13. Harshman, R.A. and Lundy, M.E. (1984a): The PARAFAC model for three-way factor analysis and multidimensional scaling. In: Research methods in multimode data analysis, Law, H.G. et al. (eds.). 122–214, Praeger, New York.Google Scholar
  14. Harshman, R.A. and Lundy, M.E. (1984b): Data preprocessing an the extended PARAFAC model. In: Research methods in mnitimode data analysis, Law, H.G. et al. (eds.), 216–284, Praeger, New York.Google Scholar
  15. Harshman, R.A. and Lundy, M.E. (1994): PARAFAC: Parallel factor analysis. Computational Statistics and Data Analysis, 18, 39–72.MATHCrossRefGoogle Scholar
  16. Hayashi, C. and Hayashi, F. (1982): A new algorithm to solve PARAFAC-model. Behaviormetrika. 11, 49–60.CrossRefGoogle Scholar
  17. Heiser, W.J. and Kroonenberg, P.M. (1994): Dimensionwise fitting in Parafac-Candecomp with missing data and constrained parameters. Unpublished manuscript, Department of Data Theory, Leiden University, Leiden.Google Scholar
  18. Kettenring, J.R. (1983): Components of interaction in analysis of variance models with no replications. In: Contributions to statistics: Essays in honor of Norman L. Johnson, Sen, P.K. (ed.), North-Holland, Amsterdam. Kiers, H.A.L. and Krijnen, W.P. (1991): An efficient algorithm for PARAFAC of three-way data with large numbers of observation units. Psychometrika, 56, 147–152.Google Scholar
  19. Krijnen, W.P. (1993): The analysis of three-way arrays by constrained PARAFAC methods, DSWO Press, Leiden.Google Scholar
  20. Krijnen, W.P. and Kroonenberg, P.M. (submitted): Detecting degeneracy when fitting the PARAFAC model.Google Scholar
  21. Krijnen, W.P. and Ten Berge, J.M.F. (1992): A constrained PARAFAC method for positive manifold data. Applied Psychological Measurement, 16, 295–305.CrossRefGoogle Scholar
  22. Kroonenberg, P.M. (1983): Three-mode principal component analysis: Theory and applications, DSWO Press, Leiden.Google Scholar
  23. Kroonenberg, P.M. (1994): The TUCKALS line: A suite of program for three-way data analysis. Computational Statistics and Data Analysis, 18, 73–96.MATHCrossRefGoogle Scholar
  24. Kroonenberg, P.M. (1996): 3WAYPACK user’s manual, Leiden University, Leiden.Google Scholar
  25. Kroonenberg, P.M. and De Leeuw, J. (1980): Principal component analysis of three-mode data by means of alternating least squares algorithms. Psychometrika, 45, 69–97.MathSciNetMATHCrossRefGoogle Scholar
  26. Kruskal, J.B., Harshman, R.A., arid Lundy, M.E. (1989): How 3-MFA can cause degenerate PARAFAC solutions, among other relationships. In: Mulliway data analysis, Coppi, R. and Bolasco, S. (eds.), 115–122, Elsevier, Amsterdam.Google Scholar
  27. Lawson, C.L. and Hanson, R.J. (1974): Solving least squares problems, Prentice Hall, Englewood Cliffs, NJ. Leurgans, S.E. and Ross, R.T. (1992): Multilinear models: Application in spectroscopy (with discussion). Statistical Science, 7, 289–319.Google Scholar
  28. Mayekawa, S.-I. (1987). Maximum likelihood solution to the PARAFAC model. Behaviormetrika, 21, 45–63.CrossRefGoogle Scholar
  29. Möcks, J. (1988): Decomposing event-related potentials: A new topographic components model. Biological Psychology, 26, 129–215.CrossRefGoogle Scholar
  30. Paatero, P. and Tapper, U. (1994): Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values. Environmetrics, 5, 111–126.CrossRefGoogle Scholar
  31. Paatero, P. (1995): User’s guide for positive matrix factorization programs PMF2.ExE and PMF3.EXE, Department of Physics, University of Helsinki.Google Scholar
  32. Pham, T.D. and Möcks, J. (1992). Beyond prinicpal component analysis: A trilinear decomposition model and least squares estimation. Psychometrika, 57, 203–215.MathSciNetMATHCrossRefGoogle Scholar
  33. Sands, R. and Young, F.W. (1980): Component models for three-way data: ALSCOMP3, an alternative least squares algorithm with optimal scaling features. Psychometrika, 45, 39–67.MathSciNetMATHCrossRefGoogle Scholar
  34. Smilde, A.K., Van der Graaf, P.H., and Doornbos, D.A. (1990): Multivariate calibration of reversed-phase chromatographic systems. Some designs based on three-way data analysis. Analytica Chimica.Acta, 235, 41–51.CrossRefGoogle Scholar
  35. Ten Berge, J.M.F. (1986): Three notes on three-way analysis. Paper presented at the Workshop on TUCKALS and PARAFAC, Leiden University, July 2.Google Scholar
  36. Ten Berge, J.M.F., Kiers, H.A.L., and Krijuen, W.P. (1993): Computational solutions for the problem of negative saliences and nonsymmetry in INDSCAL. Journal of Classification, 10, 115–124.MATHCrossRefGoogle Scholar
  37. Tucker, L.R. (1966): Some mathematical notes on three-mode factor analysis. Psychometrika, 37, 279–311.CrossRefGoogle Scholar
  38. Tucker, L.R. (1972): Relations between multidimensional scaling and three-mode factor analysis. Psychometrika, 37, 3–27.MathSciNetMATHCrossRefGoogle Scholar
  39. Van der Kloot, W.A. and Kroonenberg, P.M. (1985): External analysis with three-mode principal component analysis. Psychometrika,.Google Scholar
  40. Van Eeuwijk, F.A., Denis, J.-B., Rang, M.S. (1995): Incorporating additional information on genotypes and environments in models for two-way genotype by environment tables. In: Genotype by environment interaction: New perspectives, M.S. Rang and H.G. Gaugh Jr. (eds.), CRC-Press, Boca Raton, USA.Google Scholar
  41. Yoshizawa, T. (1988): Singular value decomposition of multiarray data and its applications. In Recent developments in clustering and data analysis, C. Hayashi et al. (eds.), Academic Press, New York.Google Scholar

Copyright information

© Springer Japan 1998

Authors and Affiliations

  • Pieter M. Kroonenberg
    • 1
  • Willem J. Heiser
    • 2
  1. 1.Department of EducationLeiden UniversityLeidenThe Netherlands
  2. 2.Department of Data TheoryLeiden UniversityLeidenThe Netherlands

Personalised recommendations