, Volume 31, Issue 3, pp 351–368 | Cite as

Factor analysis by minimizing residuals (minres)

  • Harry H. Harman
  • Wayne H. Jones


This paper is addressed to the classical problem of estimating factor loadings under the condition that the sum of squares of off-diagonal residuals be minimized. Communalities consistent with this criterion are produced as a by-product. The experimental work included several alternative algorithms before a highly efficient method was developed. The final procedure is illustrated with a numerical example. Some relationships of minres to principal-factor analysis and maximum-likelihood factor estimates are discussed, and several unresolved problems are pointed out.


Public Policy Experimental Work Factor Loading Efficient Method Statistical Theory 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Bargmann, R.Factor analysis program for 7090 preliminary version. IBM Research Center, Inter. Doc. 28–126, October 1963.Google Scholar
  2. [2]
    Boldt, R. F. Factoring to fit off diagonals.Psychometrika (in press).Google Scholar
  3. [3]
    Comrey, A. L. The minimum residual method of factor analysis.Psychol. Reports, 1962,11, 15–18.Google Scholar
  4. [4]
    Eckart, C. and Young, G. The approximation of one matrix by another of lower rank.Psychometrika, 1936,1, 211–18.Google Scholar
  5. [5]
    Finkel, R. W.The method of resultant descents for the minimization of an arbitrary function. Paper 71, preprint of paper presented at 14th National Meeting of Association of Computing Machinery, 1959.Google Scholar
  6. [6]
    Gengerelli, J. A. A simplified method for approximating multiple regression coefficients.Psychometrika, 1948,13, 135–46.Google Scholar
  7. [7]
    Harman, H. H.Modern factor analysis. Chicago, Ill.: Univ. Chicago Press, 1960.Google Scholar
  8. [8]
    Horst, P. A method of factor analysis by means of which all coordinates of the factor matrix are given simultaneously.Psychometrika, 1937,2, 225–36.Google Scholar
  9. [9]
    Householder, A. S. and Young, G. Matrix approximation and latent roots.Amer. math. Monthly, 1938,45, 165–71.Google Scholar
  10. [10]
    Hotelling, H. Analysis of a complex of statistical variables into principal components.J. educ. Psychol., 1933,24, 417–41, 498–520.Google Scholar
  11. [11]
    Howe, W. G.Some contributions to factor analysis. Report No. ORNL-1919. Oak Ridge, Tenn.: Oak Ridge National Laboratory, 1955. (Ph.D. Thesis, Univ. North Carolina.)Google Scholar
  12. [12]
    Jöreskog, K. G.Testing a simple structure hypothesis in factor analysis. Res. Bull. RB-65-1. Princeton, N. J.: Educ. Test. Serv., January 1965.Google Scholar
  13. [13]
    Keller, J. B. Factorization of matrices by least-squares.Biometrika, 1962,49, 239–42.Google Scholar
  14. [14]
    Lawley, D. N. A modified method of estimation in factor analysis and some large sample results.Uppsala Symposium on Psychological Factor Analysis. Uppsala: Almqvist & Wiksell, 1953, 35–42.Google Scholar
  15. [15]
    Shah, B. V., Buehler, R. J., and Kempthorne, O. Some algorithms for minimizing a function of several variables.J. Society for Industrial and Applied Mathematics, 1964,12, 74–92.Google Scholar
  16. [16]
    Spang, H. A., III. A review of minimization techniques for nonlinear functions.SIAM Review, 1962,4, 343–65.Google Scholar
  17. [17]
    Thurstone, L. L.Multiple-factor analysis. Chicago, Ill.: Univ. Chicago Press, 1947.Google Scholar
  18. [18]
    Thurstone, L. L. A method of factoring without communalities.1954 Invitational Conference on Testing Problems. Princeton, N. J.: Educ. Test. Serv., 1955, 59–62, 64–66.Google Scholar
  19. [19]
    Whittaker, E. and Robinson, G.The calculus of observations. London: Blackie & Son, 1944.Google Scholar
  20. [20]
    Whittle, P. On principal components and least square methods of factor analysis.Skand. Aktuar., 1952,35, 223–39.Google Scholar

Copyright information

© Psychometric Society 1966

Authors and Affiliations

  • Harry H. Harman
    • 1
  • Wayne H. Jones
    • 1
  1. 1.System Development CorporationUSA

Personalised recommendations