Statistics and Computing

, Volume 8, Issue 2, pp 125–133 | Cite as

On a differential equation approach to the weighted orthogonal Procrustes problem



The weighted orthogonal Procrustes problem, an important class of data matching problems in multivariate data analysis, is reconsidered in this paper. It is shown that a steepest descent flow on the manifold of orthogonal matrices can naturally be formulated. This formulation has two important implications: that the weighted orthogonal Procrustes problem can be solved as an initial value problem by any available numerical integrator and that the first order and the second order optimality conditions can also be derived. The proposed approach is illustrated by numerical examples.

constrained regression Procrustes rotation projected gradient optimality condition 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Bellman, R. (1970) Introduction to Matrix Analysis. 2nd edn. New York: McGraw-Hill.Google Scholar
  2. Chu, M. T. (1994) A list of matrix flows with applications. Fields Institute Communications, 3, 87–97.Google Scholar
  3. Chu, M. T. and Driessel, K. R. (1990) The projected gradient method for least squares matrix approximations with spectral constraints. SIAM Journal of Numerical Analysis, 27, 1050–60.Google Scholar
  4. Chu, M. T. and Trendafilov, N. (1996) The orthogonally constrained regression revisited. Submitted for publication.Google Scholar
  5. Gear, C. W. (1986) Maintaining solution invariants in the numerical solution of ODEs. SIAM Journal of Sci. Stat. Comput., 7, 734–43.Google Scholar
  6. Gill, P. E., Murray, W. and Wright, M. H. (1981) Practical Optimization. Florida: Academic Press.Google Scholar
  7. Golub, G. H. and Van Loan, C. F. (1991) Matrix Computation. 2nd edn. Baltimore: The Johns Hopkins University Press.Google Scholar
  8. Gower, J. C. (1984) Multivariate analysis: Ordination, multidimensional scaling and allied topics. In Emlyn Lloyd (ed.) Handbook of Applicable Mathematics, Volume IV: Statistics, part B. New York: John Wiley & Sons.Google Scholar
  9. Guillemin, V. and Pollack, A. (1974) Differential Topology. New Jersey: Prentice-Hall Inc.Google Scholar
  10. Helmke, U. and Moore, J. B. (1994) Optimization and Dynamical Systems. London: Springer Verlag.Google Scholar
  11. Hirsch, M. W. and Smale, S. (1974) Differential Equations, Dynamical Systems, and Linear Algebra. London: Academic Press.Google Scholar
  12. Koschat, M. A. and Swayne, D. F. (1991) A weighted Procrustes criterion. Psychometrika, 56, 229–39.Google Scholar
  13. Mooijaart, A. and Commandeur, J. J. F. (1990) A general solution of the weighted orthonormal Procrustes problem. Psychometrika, 55, 657–63.Google Scholar
  14. Mulaik, S. A. (1972) The Foundations of Factor Analysis. New York: McGraw-Hill.Google Scholar
  15. Ortega, J. and Rheinboldt, W. (1970) Iterative Solutions of Non-linear Equations in Several Variables. London: Academic Press.Google Scholar
  16. Shampine, L. F. and Reichelt, M. W. (1997) The MATLAB ODE suite. SIAM Journal on Scientific Computing, 18, 1–22.Google Scholar
  17. ten Berge, J. M. F. (1977) Orthogonal Procrustes rotation for two or more matrices. Psychometrika, 42, 267–76.Google Scholar

Copyright information

© Kluwer Academic Publishers 1998

Authors and Affiliations

    • 1
    • 2
  1. 1.Department of MathematicsNorth Carolina State UniversityRaleighUSA
  2. 2.Computer Stochastics Laboratory, Institute of MathematicsBulgarian Academy of SciencesSofiaBulgaria

Personalised recommendations