The strong uniqueness constant in complex approximation
- 481 Downloads
Let f(x) belong to the set C(B) of continuous functions, possibly complex valued, on a compact subset, B, of a finite dimensional euclidean space. Let V be a finite dimensional subspace of C(B). v0 ε V is a strongly unique best approximation to f if (1) ‖f − v‖ ≥ ‖f − v0‖+γ ‖v − v0∥ holds for some γ(f, B, V) > 0 and all v ε V. The largest γ (= γ*) such that (1) holds is the strong uniqueness constant for f. The strong uniqueness constant has previously been determined in essentially one case, namely, the approximation of a monomial by lower degree polynomials in the real case. When f(z)=zn, B is the closed unit disc in the complex plane and V is the set of polynomials of degree at most n−1, γ*=1/n. We show that this fact is a trivial consequence of a simple result of Szász (1917). Some related results and problems are also discussed.
Unable to display preview. Download preview PDF.
- 3.Gutknecht, M. H., On complex rational approximation; Part I: The characterization problem, Computational Aspects of Complex Analysis (Eds. H. Werner, L. Wuytack, E. Ng, H. J. Bünger), D. Reidel, Dordrecht, Holland (1973), 79–101.Google Scholar
- 4.Newman, D. J., Polynomials and rational functions, Approximation Theory and Applications (Ed. Z. Ziegler), Academic Press, N. Y. (1981), 265–282.Google Scholar
- 6.Rivlin, T., The best strong uniqueness constant for a multivariate Chebyshev polynomial, J. Approx. Theory, to appear.Google Scholar
- 7.Szász, O., Über nichtnegative trigonometrische Polynome, Sitzungsb. d. Bayer. Akad. der Wiss. Math. — Phys. Kl., 1917, 307–320.Google Scholar