Roots of Polynomials Expressed in Terms of Orthogonal Polynomials

Authors:
David Day;Louis Romero
Affiliations:
-;-
Venue:
SIAM Journal on Numerical Analysis
Year:
2005

Citing 0
Cited 8

Computing real roots of a polynomial in Chebyshev series form through subdivision

Applied Numerical Mathematics
Rootfinding for a transcendental equation without a first guess: polynomialization of Kepler's equation through Chebyshev polynomial expansion of the sine

Applied Numerical Mathematics
Numerical experiments on the accuracy of the Chebyshev-Frobenius companion matrix method for finding the zeros of a truncated series of Chebyshev polynomials

Journal of Computational and Applied Mathematics
Computing the zeros of a Fourier series or a Chebyshev series or general orthogonal polynomial series with parity symmetries

Computers & Mathematics with Applications
Chebyshev expansion on intervals with branch points with application to the root of Kepler's equation: A Chebyshev-Hermite-Padé method

Journal of Computational and Applied Mathematics
Simultaneous confidence bands for all contrasts of three or more simple linear regression models over an interval

Computational Statistics & Data Analysis
Sturm root counting using chebyshev expansion

ACM Communications in Computer Algebra
Stability of rootfinding for barycentric Lagrange interpolants

Numerical Algorithms

Quantified Score

Hi-index	0.01

Visualization

Abstract

A technique is presented for determining the roots of a polynomial $p(x)$ that is expressed in terms of an expansion in orthogonal polynomials. The roots are expressed as the eigenvalues of a nonstandard companion matrix $B_n$ whose coefficients depend on the recurrence formula for the orthogonal polynomials, and on the coefficients of the orthogonal expansion. Some questions on the numerical stability of the eigenvalue problem to which they give rise are discussed. The problem of finding the roots of a transcendental function $f(x)$ can be reduced to the problem considered by approximating $f(x)$ by a Chebyshev polynomial. We illustrate the effectiveness of this convert-to-Chebyshev strategy by solving several transcendental equations using this plus our new algorithm. We analyze the numerical stability through both linear algebra theory and numerical experiments and find that this method is very well conditioned.