On Weighted L p -Convergence of Certain Lagrange Interpolation (Q4027749)

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 126056
Language Label Description Also known as
default for all languages
No label defined
    English
    On Weighted L p -Convergence of Certain Lagrange Interpolation
    scientific article; zbMATH DE number 126056

      Statements

      On Weighted L p -Convergence of Certain Lagrange Interpolation (English)
      0 references
      0 references
      1 March 1993
      0 references
      weighted \(L^ p\)-convergence
      0 references
      Jacobi polynomial
      0 references
      Lagrange polynomial
      0 references
      Let \(w(x):=w^{(\alpha,\beta)}(x)=(1-x)^ \alpha (1+x)^ \beta\) and let \[ x:\;-1\leq x_ n<x_{n-1}<\dots<x_ 1\leq 1 \tag{1} \] be the zeros of the Jacobi polynomial with weight \(w^{(\alpha,\beta)}(x)\). Let \(L_ n(f,w,x)\) denote the Lagrange polynomial on the nodes (1). \textit{P. Erdős} and \textit{E. Feldheim} [C. R. Acad. Sci., Paris 203, 913-915 (1936; Zbl 0015.25203)] had shown that if \(f\in C[-1,1]\) then \[ \lim_{n\to\infty} \int_{-1}^ 1 {1\over {\sqrt{1-x^ 2}}} | L_ n(f,w^{-{1\over 2},-{1\over 2})}(x))-f(x)|^ p dx=0, \qquad p>0.\tag{2} \] \textit{A. Bellen} [J. Approximation Theory 33, 85-95 (1982; Zbl 0484.41003)] raised the following problem: Find a matrix \(Y\) of additional nodes \(Y\): \(-1\leq y_{m(n)}<y_{m(n)-1}<\dots<y_ 1\leq 1\), \(x_ i\neq y_ j\), \(\forall i,j\), where \(m(n)=an+b\), \(a\), \(b\) real, \(a>0\), such that \(L_{n+m(n)}(f,x):=L_{n+m(n)}(f,X\cup Y,x)\) satisfies the relation \[ \lim_{n\to\infty} \int_{-1}^ 1 w(x)[L_{n+m(n)}(f;x)-f(x)]^ 2 dx=0\tag{3} \] for some class of functions \(f\) on \([-1,1]\). He showed that if \(X\) consists of zeros of \((1-x^ 2)U_{n-1}(x)\) and \(Y\) is formed by zeros of \(T_ n(x)\), then (3) holds with \(w(x)={1\over{\sqrt{1-x^ 2}}}\) when \(f\in\text{Lip }\gamma\), \(\gamma>{1\over 2}\). Here the author improves this result in an elegant way by showing that if \(f\in C[-1,1]\), then \[ \lim_{n\to\infty} \int_{-1}^ 1 {1\over {\sqrt{1-x^ 2}}}:\;| L_{2n+1}(f;x)- f(x)|^ p dx=0, \qquad p>0.\tag{4} \] .
      0 references

      Identifiers