Abstract
In a linear Gauss–Markov model, the parameter estimates from BLUUE (Best Linear Uniformly Unbiased Estimate) are not robust against possible outliers in the observations. Moreover, by giving up the unbiasedness constraint, the mean squared error (MSE) risk may be further reduced, in particular when the problem is ill-posed. In this paper, the α-weighted S-homBLE (Best homogeneously Linear Estimate) is derived via formulas originally used for variance component estimation on the basis of the repro-BIQUUE (reproducing Best Invariant Quadratic Uniformly Unbiased Estimate) principle in a model with stochastic prior information. In the present model, however, such prior information is not included, which allows the comparison of the stochastic approach (α-weighted S-homBLE) with the well-established algebraic approach of Tykhonov–Phillips regularization, also known as R-HAPS (Hybrid APproximation Solution), whenever the inverse of the “substitute matrix” S exists and is chosen as the R matrix that defines the relative impact of the regularizing term on the final result.
Similar content being viewed by others
References
Arsenin VY, Krianev AV (1992) Generalized maximum likelihood method and its application for solving ill-posed problems. In: Tykhonov A (ed) Ill-Posed problems in natural sciences. TVP Science Publishers, Moscow, pp 3–12
Bouman J (1998) Quality of regularization methods, DEOS Report No. 98.2, Delft University Press, NL
Engl H (1993) Regularization methods for the stable solution of inverse problems. Surv Math Ind 3:371–143
Engl H, Hanke M, Neubauer A (1996) Regularization of Inverse Problems. Kluwer, Dordrecht
Farebrother RW (1975) The minimum mean squared error linear estimator and ridge regression. Technometrics 17:127–128
Farebrother RW (1976) Further results on the mean squared error of ridge regression. J R Stat Soc B 38:248–250
Farebrother RW (1978) Partitioned ridge regression. Technometrics 20:121–122
Frommer A, Maas P (1999) Fast CG-based methods for Tykhonov–Phillips regularization. SIAM J Sci Comput 20:1831–1850
Golub GH, Heath M, Wahba G (1979) Generalized cross-validation as a method for choosing a good ridge parameter. Technometrics 21:215–223
Golub GH, Hansen PC, O’Leary DP (1999) Tykhonov regularization and total least squares. SIAM J Matrix Anal Appl 21:185–194
Grafarend EW, Schaffrin B (1993) Ausgleichungsrechnung in linearen Modellen. Bibliographisches Institut, Mannheim
Groetsch CW (1984) The theory of Tykhonov regularization for Fredholm equations of the first Kind, Pitman: London
Groetsch CW (1993) Inverse problems in the mathematical sciences. Vieweg, Wiesbaden
Gunst RF, Mason RL (1977) Biased estimation in regression: An evaluation using mean squared error. J Am Stat Assoc 72:616–628
Gunst RF, Mason RL (1980) Regression analysis and its application. Marcel Dekker, New York
Hanke M, Hansen P (1993) Regularization methods for large scale problems. Surv Math Ind 3:253–315
Hansen PC (1992) Analysis of discrete ill-posed problems by means of the L-curve. SIAM Rev 34:561–580
Hansen PC (1994) Regularization tools. A MATLAB package for analysis and solution of discrete ill-posed problems, Numer Algorithms 6:1–35
Hansen PC (1997) Rank-deficient and discrete ill-posed problems. Numerical aspects of linear inversion. SIAM, Philadelphia PA
Hansen PC, O’Leary DP (1993) The use of the L-curve in the regularization of discrete ill-posed problems. SIAM J Sci Comput 14:1487–1503
Hemmerle WJ (1975) An explicit solution for generalized ridge regression, Technometrics 17:309–314
Hoerl AE, Kennard RW (1970a) Ridge regression: Biased estimation for nonorthogonal problems. Technometrics 12:55–67
Hoerl AE, Kennard RW (1970b) Ridge regression: Applications to nonorthogonal problems. Technometrics 12:69–82
Hoerl AE, Kennard RW, Baldwin KF (1975) Ridge regression: Some simulations analysis. Commun Stat 4:105–123
Hoerl RW (1985) Ridge analysis 25 years later. Am Stat 39:186–192
Ilk KH (1986) On the regularization of ill-posed problems. In: Proceedings of the international symposium on Figure and Dynamics of the Earth, Moon and planets, Prague, pp 365–383
Ilk KH (1993) Regularization for high-resolution gravity field recovery by future satellite techniques. In: Anger G, Gorenflo R, Jochmann H, Moritz H, Webers W (eds). Inverse problems: principles and applications in geophysics, technology and medicine. Akademie-Verlag, Berlin, pp 189–214
Iz HB (1987) An algorithmic approach to crustal deformation analysis, OSU-Report No. 382, Department of Geodetic Science and Surveying. The Ohio State University, Columbus, OH, USA
Johnston PR, Gulrajani RM (1997) A new method for regularization parameter determination in the inverse problem of electrocardiography. IEEE Trans Biomed Eng 44:19–39
Johnston PR, Gulrajani RM (2002) An analysis of the zero-crossing method for choosing regularization parameters. SIAM J. Sci Comput 24:428–442
Kacirattin S, Sakalloglu S, Akdeniz F (1998) Mean squared error comparisons of the modified ridge regression estimator and the restricted ridge regression 27:131–138
Koch KR (1990) Bayesian inference with geodetic applications. Springer, Berlin
Koch KR, Kusche J (2002) Regularization of geopotential determination from satellite data by variance components. J Geod 76:259–268
Kusche J, Klees R (2002) Regularization of gravity field estimation from satellite gravity gradients. J. Geod 76:359–368
Levenberg K (1944) A method for the solution of certain problems in least-squares. Q Appl Math 2:164–168
Marquardt WD (1963) An algorithm for least-squares estimation of nonlinear parameters. SIAM J Appl Math 11:431–441
Marquardt WD (1970) Generalized inverses, ridge regression, biased linear estimation, and nonlinear estimation. Technometrics 12:591–612
Marquardt WD (1974) Discussion No.2. Technometrics 16:189–192
Marquardt WD, Snee RD (1975) Ridge regression in practice. Am Stat 29:3–20
Mayer LS, Willke TA (1973) On biased estimation in linear models. Technometrics 15:497–508
Menke W (1989) Geophysical data analysis: discrete inverse theory (revised edition). Academic Press, San Diego
Morozov VA (1984) Methods for solving incorrectly posed problems. Springer, New York
O’Sullivan F (1986) A statistical perspective on ill-posed inverse problems. Statist Sci 1:502–527
Phillips DL (1962) A technique for the numerical solution of certain integral equations of the first kind. J Ass Comput Mach 9:84–96
Rao CR (1975) Simultaneous estimation of parameters in different linear models and applications to biometric problems. Biometrics 31:545–554
Reginska T (1996) A regularization parameter in discrete ill-posed problems. SIAM J Sci Comput 17:740–749
Schaffrin B (1980) Tykhonov regularization in geodesy. An example, Boll Geod Sci Aff 39:207–216
Schaffrin B (1983) Variance-covariance component estimation for multivariate repeated measurements (in German), Publ. of the German Geodetic Comm. C-282, Munich
Schaffrin B (1995) A comparison of inverse techniques: Regularization, weight estimation and homBLUP, IUGG General Assembly, IAG Sci. Meeting U7, Boulder/Colorado, July 1995
Schaffrin B (2000) Minimum mean squared error adjustment, Part I: The empirical BLE and the repro-BLE for direct observations. J Geod Soc Japan 46:21–30
Schaffrin B (2007) On penalized least-squares: Its mean squared error and a quasi-optimal weight ratio. In: Shalabh, Heumann Ch. (eds) Recent advances in linear models and related areas, Springer, Heidelberg (forthcoming)
Schaffrin B, Heidenreich E, Grafarend E (1977) A representation of the standard gravity field. Manuscr Geod 2:135–174
Schaffrin B, Middel B (1990) Robust predictors and an alternative iteration scheme for ill-posed problem. In: Vogel A. Ofoegbu Ch O, Gorenflo R, Ursin Bj (eds). Geophysical data inversion. Methods and applications. Vieweg, Braunschweig, pp 33–51
Theobald CM (1974) Generalizations of mean squared error applied to ridge regression. J R Stat Soc B 36:103–106
Tykhonov AN (1963) The regularization of incorrectly posed problem. Sov Math Doklady 4:1624–1627
Tykhonov AN, Arsenin VY (1977) Solutions of Ill-posed problems. Wiley, New York
Tykhonov AN, Bol’shakov VD, Byvshev VA, Il’inskiy AS, Neyman YuM (1977) A variational method of regularization in the adjustment of free geodetic nets. Geod Mapp Photogrammetry 19:127–139
Wang Y, Xiao T (2001) Fast realization algorithms for determining regularization parameters in linear inverse problems, Inverse Probl 17:281–291
Xu P (1992a) Determination of surface gravity anomalies using gradiometric observables. Geophys J Int 110:321–332
Xu P (1992b) The value of minimum norm estimation of geopotential fields. Geophys J Int 111:170–178
Xu P, Rummel R (1994a) A simulation study of smoothness methods in recovery of regional gravity fields. Geophys J Int 117:472–486
Xu P, Rummel R (1994b) Generalized ridge regression with applications in the determination of potential fields. Manuscr Geod 20:8–20
Xu P (1998) Truncated SVD methods for linear discrete ill-posed problems. Geophys J Int 135:505–514
Author information
Authors and Affiliations
Corresponding author
Additional information
The delay in publishing this paper is due to a number of unfortunate complications. It was first submitted as a multi-author paper in two parts. Due to some miscommunication among the original authors, it was reassigned to one of the J Geod special issues, but later reassigned at this author’s request to a standard issue of J Geod. This compounded with a difficulty to find willing reviewers to slow the process. We apologize to the author.
Rights and permissions
About this article
Cite this article
Schaffrin, B. Minimum mean squared error (MSE) adjustment and the optimal Tykhonov–Phillips regularization parameter via reproducing best invariant quadratic uniformly unbiased estimates (repro-BIQUUE). J Geod 82, 113–121 (2008). https://doi.org/10.1007/s00190-007-0162-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00190-007-0162-0