Best Approximation by Convex Functions (Q3821092)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 4094098
Language Label Description Also known as
default for all languages
No label defined
    English
    Best Approximation by Convex Functions
    scientific article; zbMATH DE number 4094098

      Statements

      Best Approximation by Convex Functions (English)
      0 references
      0 references
      1987
      0 references
      best uniform approximation
      0 references
      best convex approximations
      0 references
      The best uniform approximation of functions continuous on \([a,b]\) by the class \(K_ c\) of continuous convex functions is considered. Proof of existence and characterization of the best approximating function is presented. The result obtained in the paper is a special case of the more general result due to \textit{H. G. Burchard} announced in [Bull. Am. Math. Soc. 79, 959--963 (1973; Zbl 0269.41010)], for which no complete proof has been published. The following result is established: Every \(f\in C[a,b]\) has a best convex approximation in \(K_ c\). A convex function \(g\) is a best approximation to \(f\in C[a,b]\setminus K_ c\) if and only if there exists an interval \([\alpha,\beta]\subseteq [a,b]\) on which \(g\) is a linear polynomial, and a point \(\tau\in (\alpha,\beta)\) such that \(f(\alpha)-g(\alpha)=f(\beta)-g(\beta)=-\| f-g\|_{\infty}\) and \(f(\tau)-g(\tau)=\| f-g\|_{\infty}.\) Furthermore, all best convex approximations to \(f\) agree on \([a,\beta]\).
      0 references

      Identifiers