Hostname: page-component-77c89778f8-cnmwb Total loading time: 0 Render date: 2024-07-17T02:22:49.151Z Has data issue: false hasContentIssue false
  • English
  • Français

The extension of generalized cross-validation to a multi-parameter class of estimators

Published online by Cambridge University Press:  17 February 2009

D. M. O'Brien  and
J. N. Holt
D. M. O'Brien
Affiliation:
Department of Mathematical Physics, University of Adelaide, Adelaide, South Australia 5000
J. N. Holt
Affiliation:
Department of Mathematics, University of Queensland, St. Lucia, Queensland 4067
Rights & Permissions [Opens in a new window]

Abstract

Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

The method of generalized cross-validation (GCV) provides a good value for the “ridge” regularization parameter for an ill-conditioned linear system, such as the system produced by discretization of a Fredholm integral equation of the first kind. In this note we apply GCV to a wider class of estimators than the one parameter ridge estimators. We observe that the expected values of the parameter mean-square error, the predictive mean-square error, and the GCV function are simultaneously minimized over this new class, so we accept the minimizer of the GCV function as the best computable estimator. We present a simple algorithm for computing this estimator from the data, so that a numerical search is not needed.

Type
Research Article
Information
The ANZIAM Journal , Volume 22 , Issue 4 , April 1981 , pp. 501 - 514
Copyright
Copyright © Australian Mathematical Society 1981

References

[1]Craven, P. and Wahba, G., “Smoothing noisy data with spline functions. Estimating the correct degree of smoothing by the method of generalized cross-validation”, Numer. Math. 31 (1979), 377–403.CrossRefGoogle Scholar
[2]Golub, G. H., Heath, M. and Wahba, G., “Generalized cross-validation as a method for chossing a good ridge parameter”, Technometrics 21 (1979), 215–223.CrossRefGoogle Scholar
[3]Hardy, G. H., Littlewood, J. E. and Pólya, G., Inequalities (Cambridge University Press, 2nd edition, 1952), Chapter X.Google Scholar
[4]de Hoog, F. R., “Review of Fredliolm equations of the first kind”, in The application and numerical solution of integral equations (eds. Anderssen, R. S., de Hoog, F. R., and Lukas, M. A.) (Sijthoff and Noordhoff, The Netherlands, 1980).Google Scholar
[5]Lukas, M. A., “Regularization”, in The application and numerical solution of integral equations (eds. Anderssen, R. S., de Hoog, F. R., and Lukas, M. A., Sijthoff and Noordhoff, The Netherlands, 1980).Google Scholar
[6]Mallows, C. L., “Some comments on Cp”, Technometrics 15 (1973), 661–675Google Scholar
[7]Tihonov, A. N., “Solution of incorrectly formulated problems and the method of regularization”, Soviet Math. Dokl. 4(1963), 1035–1038.Google Scholar
[8]Wahba, G., “The approximate solution of linear opeartor equations when the data are noisy”, SIAM J. Num. Anal. 14 (1977), 651–667.CrossRefGoogle Scholar