Abstract
Generalized cross-validation (GCV) is a popular tool for specifying the tuning parameter in linear regression model or equivalently the regularization parameter in Tikhonov regularization. In this work, we are concerned with the estimation and minimization of the GCV function by using a combination of an extrapolation procedure and a statistical approach. In particular, we derive families of estimates for the GCV function. By minimizing the estimated GCV function over a grid of values, a GCV estimate of the regularization parameter is achieved. We present several numerical examples to illustrate the effectiveness of the derived families of estimates for approximating the regularization parameter for several linear discrete ill-posed problems.
Similar content being viewed by others
References
Androulakis, E., Koukouvinos, C., Mylona, K.: Tuning parameter estimation in penalized least squares methodology. Commun. Stat. Simul. Comput. 40(9), 1444–1457 (2011)
Bouhamidi, A., Enkhbat, R., Jbilou, K.: Conditional gradient Tikhonov method for a convex optimization problem in image restoration. J. Comput. Appl. Math. 255, 580–592 (2014)
Brezinski, C.: Error estimates for the solution of linear systems. SIAM J. Sci. Comput. 21, 764–781 (1999)
Brezinski, C., Fika, P., Mitrouli, M.: Estimations of the trace of powers of positive self-adjoint operators by extrapolation of the moments. Electron. Trans. Numer. Anal. 39, 144–155 (2012)
Brezinski, C., Rodriguez, G., Seatzu, S.: Error estimates for the regularization of least squares problems. Numer. Algo. 51, 61–76 (2009)
Craven, P., Wahba, G.: Smoothing noisy data with spline functions. Numer. Math. 31, 377–403 (1979)
Fenu, C., Reichel, L., Rodriguez, G.: GCV for Tikhonov regularization via global Golub–Kahan decomposition. Numer. Linear Algebra Appl. 23, 467–484 (2016)
Fenu, C., Reichel, L., Rodriguez, G., Sadok, H.: GCV for Tikhonov regularization by partial singular value decomposition. BIT Numer. Math. (2017). https://doi.org/10.1007/s10543-017-0662-0
Fika, P., Koukouvinos, C.: Stochastic estimates for the trace of functions of matrices via Hadamard matrices. Commun. Stat. Simul. Comput. 46(5), 3503–3941 (2017)
Fika, P., Mitrouli, M.: Estimation of the bilinear form \(y^*f(A)x\) for Hermitian matrices. Linear Algebra Appl. 502, 140–158 (2016)
Fika, P., Mitrouli, M., Roupa, P.: Estimates for the bilinear form \(x^TA^{-1}y\) with applications to linear algebra problems. Electron. Trans. Numer. Anal. 43, 70–89 (2014)
Fika, P., Mitrouli, M., Roupa, P.: Estimating the diagonal of matrix functions. Math. Methods Appl. Sci. 41(3), 1083–1088 (2018)
Gazzola, S., Novati, P., Russo, M.R.: On Krylov projection methods and Tikhonov regularization. Electron. Trans. Numer. Anal. 44, 83–123 (2015)
Golub, G.H., Heath, M., Wahba, G.: Generalized cross-validation as a method for choosing a good ridge parameter. Technometrics 21, 215–223 (1979)
Golub, G.H., Meurant, G.: Matrices, Moments and Quadrature with Applications. Princeton University Press, Princeton (2010)
Golub, G.H., von Matt, U.: Generalized cross-validation for large-scale problems. J. Comput. Graph. Stat. 6, 1–34 (1997)
Hansen, P.C.: Regularization tools Version 4.0 for MATLAB 7.3. Numer. Algorithms 46, 189–194 (2007)
Higham, N.J.: Functions of Matrices: Theory and Computation. SIAM, Philadelphia, PA (2008)
Hutchinson, M.F.: A stochastic estimator of the trace of the influence matrix for Laplacian smoothing splines. Commun. Stat. Simul. 19, 433–450 (1990)
Reichel, L., Rodriguez, G., Seatzu, S.: Error estimates for large-scale ill-posed problems. Numer. Algorithms 51(3), 341–361 (2009)
Acknowledgements
The authors are grateful to the reviewers of this paper whose valuable remarks improved significantly this work. The work of the second author (P. R.) was implemented with a scholarship from State Scholarships Foundation (IKY) which was funded by the Act “Scholarship Program for postgraduate studies of a second cycle of studies” from resources of the OP “Human Resources Development, Education and Lifelong Learning”, 2014-2020 with the co-financing of the European Social Fund (ESF) and the Greek State.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Mitrouli, M., Roupa, P. Estimates for the generalized cross-validation function via an extrapolation and statistical approach. Calcolo 55, 24 (2018). https://doi.org/10.1007/s10092-018-0266-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10092-018-0266-3