Skip to main content
SpringerLink
Log in

Implementing partial least squares

  • Papers
  • Published:
Statistics and Computing Aims and scope Submit manuscript

Abstract

Partial least squares (PLS) regression has been proposed as an alternative regression technique to more traditional approaches such as principal components regression and ridge regression. A number of algorithms have appeared in the literature which have been shown to be equivalent. Someone wishing to implement PLS regression in a programming language or within a statistical package must choose which algorithm to use. We investigate the implementation of univariate PLS algorithms within FORTRAN and the Matlab (1993) and Splus (1992) environments, comparing theoretical measures of execution speed based on flop counts with their observed execution times. We also comment on the ease with which the algorithms may be implemented in the different environments. Finally, we investigate the merits of using the orthogonal invariance of PLS regression to ‘improve’ the algorithms.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Brown, P. J. (1990) Partial least squares in perspective. Analytical Proceedings of the Royal Society of Chemistry, 27, 303–306.

    Google Scholar 

  • CAMO A/S (1987) UNSCRAMBLER users guide, version 2. Trondheim, Norway.

    Google Scholar 

  • Chambers, J. M. (1991) Data management in S. Technical Report. Statistical Research Report, No. 99 AT&T Bell Laboratories, Murray Hill, New Jersey.

    Google Scholar 

  • Denham, M. C. (1991) Multivariate calibration in infrared spectroscopy. Ph.D. Thesis. University of Liverpool.

  • Dongarra, J. J., Bunch, J. R., Moler, C. B. and Stewart, G. W. (1979) INPACK User's Guide. SIAM, Philadelphia.

    Google Scholar 

  • Dongarra, J. J., DuCroz, J., Duff, I. and Hammarling, S. (1988) A set of level 3 basic linear algebra subprograms. Technical Report. Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439.

    Google Scholar 

  • Dongarra, J. J., DuCroz, J., Hammarling, S. and Hanson, R. (1988) An extended set of fortran basic linear algebra subprograms. ACM Transactions in Mathematical Software, 14, 1–32.

    Google Scholar 

  • Genz, A., Jones, C., Luo, D. and Prenzel, T. (1991) Fast Givens goes slow in MATLAB. SIGNUM Newsletter, 26(2), 11–16.

    Google Scholar 

  • Golub, G. H. and Van Loan, C. F. (1989) Matrix computations, 2nd edn. Johns Hopkins University Press, Baltimore, MD.

    Google Scholar 

  • Helland, I. S. (1988) On the structure of partial least squares regression. Communications in Statistics—Elements of Simulation and Computation, 17, 581–607.

    Google Scholar 

  • Lawson, C., Hanson, R., Kincaid, D. and Krogh, F. (1979) Basic linear algebra subprograms for Fortran usage. ACM Transactions in Mathematical Software, 5, 308–325.

    Google Scholar 

  • Martens, H. and Jensen, S. A. (1983) Partial least squares regression: a new two-stage NIR calibration method. In Progress in Cereal Chemistry and Technology 5a, Proceedings, 7th World Cereal and Bread Congress, Prague, June 1982, ed. J. Holas and J. Kratochvil, pp. 607–647. Elsevier, Amsterdam.

    Google Scholar 

  • Martens, H. and Næs, T. (1989) Multivariate Calibration. Wiley, Chichester.

    Google Scholar 

  • MATLAB (1993) Matlab User Manual. Mathworks, Inc.

  • Numerical Algorithms Group (1990) NAG Fortran Library Manual, Mk. 14. NAG, Oxford.

    Google Scholar 

  • Rogers, C. A. (1987) A genstat macro for partial least squares analysis with cross-validation assessment of model dimensionality. Genstat Newsletter, 18.

  • Splus (1992) Splus for Sun workstations, version 3.1. Statistical Sciences Inc., Oxford.

    Google Scholar 

  • Stone, M. and Brooks, R. (1990) Continuum regression: crossvalidated sequentially-constructed prediction embracing ordinary least squares, partial least squares, and principal components regression. Journal of the Royal Statistical Society, Series B, 52, 237–269.

    Google Scholar 

  • WATCOM (1989) Maple Reference Manual. WATCOM Publications, Waterloo, Ontario.

    Google Scholar 

  • Wold, H. (1966) Nonlinear estimation by iterative least squares procedures. In Research papers in statistics: Festschrift for J. Neyman, ed. F. N. David, pp. 411–444. Wiley, New York.

    Google Scholar 

  • Wold, H. (1973) Nonlinear iterative partial least squares (NIPALS) modelling: Some current developments. In Multivariate Analysis III, ed. P. Krishnaiah, pp. 383–407. Academic Press, New York.

    Google Scholar 

  • Wold, H. (1982) Soft modeling: The basic design and some extensions. In Systems under indirect observation: causality-structure-prediction, eds K. G. Jöreskog and H. Wold, Vol. II, Ch. 1, pp. 1–54. North-Holland, Amsterdam.

    Google Scholar 

Download references

Author information

Author notes

  1. M. C. Denham

    Present address: Department of Applied Statistics, University of Reading, Harry Pitt Building, Whiteknights Road, PO Box 240, RG6 2FN, Reading, USA

Authors and Affiliations

  1. Department of Statistics and Computational Mathematics, University of Liverpool, L69 3BX, Liverpool, UK

    M. C. Denham

Authors
  1. M. C. Denham
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

About this article

Cite this article

Denham, M.C. Implementing partial least squares. Stat Comput 5, 191–202 (1995). https://doi.org/10.1007/BF00142661

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF00142661

Keywords

Search

Navigation