Abstract
This paper describes a variant of the Gauss-Newton-Hartley algorithm for nonlinear least squares, in which aQR implementation is used to solve the linear least squares problem. We follow Grey's idea of updating variables at intermediate stages of the orthogonalization. This technique, applied in partitions identified with known or suspected spectral lines, appears to be especially suited to the analysis of spectroscopic data. We suggest that this algorithm is an attractive candidate for the optimization role in Ekenberg's interactive computer graphics curve fitting program.
This is a preview of subscription content, log in via an institution to check access.
References
R. H. Barham and W. Drane,An algorithm for least squares estimation of nonlinear parameters when some of the parameters are linear, Technometrics 14 (1972), 757–766.
N. A. Broste,A comparison of an orthonormal optimization program and conjugate gradients, Carnegie Mellon Univ.: Pittsburgh 1968, Univ. Microfilm #69-7898.
L. W. Cornwell, R. J. Pegis, A. K. Rigler, and T. P. Vogl,Grey's method for nonlinear optimization, J. Opt. Soc. Am. 63 (1973), 576–581.
G. Dahlquist and Å. Björck,Numerical Methods, Prentice-Hall, Englewood Cliffs, 1974.
B. Ekenberg,Curve fitting by use of graphic display, BIT 15 (1975), 385–393.
G. H. Golub and V. Pereyra,The differentiation of pseudo-inverses and nonlinear least squares problems whose variables separate, SIAM J. Num. Anal. 10 (1973), 413–432.
W. B. Gragg, R. J. Leveque, and J. A. Trangenstein,Numerically stable methods for updating regressions, J. Am. Stat. Soc. 74 (1979), 161–168.
D. S. Grey,Aberration theories for semi-automatic lens design by electronic computers: I.Preliminary remarks, II.A specific computer program, J. Opt. Soc. Am. 53 (1963), 672–676, 677–680.
H. O. Hartley,The modified Gauss-Newton method for fitting of nonlinear regression functions by least squares, Technometrics 3 (1961), 269–280.
R. Hooke and R. A. Jeeves,Direct search solution of numerical and statistical problems, J.A.C.M. 8 (1961), 212–229.
L. Kaufman,A variable projection method for solving separable nonlinear least squares problems, BIT 15 (1975), 49–57.
L. Kaufman and V. Pereyra,A method for separable nonlinear least squares problems with separable nonlinear equality constraints, SIAM J. Num. Anal. 15 (1978), 12–20.
C. L. Lawson and R. J. Hanson,Solving Least Square Problems, Prentice-Hall, Englewood Cliffs, 1974.
J. P. Lesnick and A. K. Rigler,Computer curve fitting of spectral absorptivity data, Westinghouse Research Laboratories Report, 67-9Cl-LH1CW-R3, AD-662, 252 (1967).
J. R. Rice,Experiments on Gram-Schmidt orthogonalization, Math. Comp. 20 (1966), 325–328.
A. K. Rigler and D. C. St. Clair,Note on Gram-Schmidt orthogonalization in regression calculations, University of Missouri-Rolla, Computer Science Report (1979).
A. Ruhe and P. Å. Wedin,Algorithms for separable non-linear least squares problems, Stanford University Computer Science Report STAN-CS-74-434 (1974).
D. C. St. Clair and A. K. Rigler,A block orthogonalization method for nonlinear regression, Computer Science and Statistics: Proceedings of the Eighth Annual Symposium on the Interface (1975), 280–287.
T. P. Vogl, L. C. Lintner and A. K. Rigler,The lens II optical design computer program for IBM OS/360, Frankford Arsenal and White Sands Missile Range, Report on Contract #DAAA-25-67-C0445 (1967).
R. E. von Holdt,Inversion of triple diagonal compound matrices, J.A.C.M. 9 (1962), 71–83.
D. D. Walling,Nonlinear least squares curve fitting when some parameters are linear, The Texas Journal of Science 20 (1968), 119–124.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Clair, D.C., Rigler, A.K. An algorithm for least squares analysis of spectroscopic data. BIT 19, 448–456 (1979). https://doi.org/10.1007/BF01931260
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01931260