On the condition number of some gram matrices arising from least squares approximation in the complex plane Article DOI:
Cite this article as: Saad, Y. Numer. Math. (1986) 48: 337. doi:10.1007/BF01389479 Summary
This paper analyses the growth of the condition number of a class of Gram matrices that arise when computing least squares polynomials in polygons of the complex plane. It is shown that if the polygon is inserted between two ellipses then the condition number of the (
n+1)×( n+1) Gram matrix is bounded from above by 4 m( n+1) 2( k) 2 where n m is the number of edges of the polygon, and k≥1 is a known ratio which is close to one if the two ellipses are close to each other. Subject Classifications AMS(MOS): 65D99 CR: G1.2
This work was supported in part by the U.S. Office of Naval Research under grant N00014-82-K-0184, in part by Dept. Of Energy under Grant AC02-81ER10996, and in part by Army Research Office under contract DAAG-83-0177
Davis, P.J.: Interpolation and Approximation. Waltham, MA: Blaisdell 1963
Gautschi, W.: On generating orthogonal polynomials. SIAM J. Sci. Stat. Comput.
, 289–317 (1982)
Gautschi, W.: Construction of Gauss-Christoffel quadrature formulas. Math. Comput.
, 251–270 (1968)
Lorentz, G.G.: Approximation of functions. New York: Holt, Rinehart & Winston 1966
Manteuffel, T.A.: An iterative method for solving nonsymmetric linear systems with dynamic estimation of parameters. Technical Report UIUCDCS-75-758, University of Illinois at Urbana-Champaign (Ph. D. dissertation) 1975
Rivlin, T.J.: The Chebyshev Polynomials. New York: Wiley 1976
Saad, Y.: Least squares polynomials in the complex plane and their use for solving sparse nonsymmetric linear systems. SIAM J. Sci. Stat. Comput. (Submitted)
Saad, Y.: Least squares polynomials in the complex plane with applications to solving sparse nonsymmetric matrix problems. Technical Report 276, Yale University, Computer Science Dept. 1983
Saad, Y.: Practical use of polynomial preconditionings for the Conjugate Gradient Method. Technical Report YALEU/DCS/RR-282, Yale University, 1983. SIAM J. Sci. Stat. Comput.
, 865–881 (1985)
Sack, R.A., Donovan, A.F.: An algorithm for Gaussian quadrature given modified moments. Numer. Math.
, 465–478 (1971)
Smolarski, D.C.: Optimum semi-iterative methods for the solution of any linear algebraic system with a square matrix. Technical Report UIUCDCS-R-81-1077, University of Illinois at Urbana-Champaign (Ph. D. Thesis) 1981
Smolarski, D.C., Saylor, P.E.: Optimum parameters for the solution of linear equations by Richardson iteration, 1982 (To appear)