Converse theorems for approximation by Bernstein polynomials in \(L_ p[0,1]\) \((1<p<\infty)\) (Q1084273): Difference between revisions
From MaRDI portal
Latest revision as of 16:29, 17 June 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Converse theorems for approximation by Bernstein polynomials in \(L_ p[0,1]\) \((1<p<\infty)\) |
scientific article |
Statements
Converse theorems for approximation by Bernstein polynomials in \(L_ p[0,1]\) \((1<p<\infty)\) (English)
0 references
1986
0 references
The paper deals with the question of characterizing those continuous functions f(x) on [0,1] that are approximable by their Bernstein polynomials \(B_ nf\) in the \(L^ p\)-norm, \(1<p<\infty\), at certain rates. The functions are characterized by means of their second \(\tau\)- modulus of smoothness. Among the results is the following Theorem. For \(1<p<\infty\), \(1/p<\alpha <1\) and \(f\in C[0,1]\), the following statements are equivalent. (1) \(\| f-B_ nf\|_ p=O(n^{-\alpha});\) (2) \(\tau_ 2(f,\Delta (\delta))_{p,p}=O(\delta^{2\alpha});\) (3) \(\tau_ 2(f,\Delta (\delta))_ p=O(\delta^{2\alpha})\) where \(\Delta (\delta)=\Delta (\delta,x)=\delta \sqrt{x(1-x)}+\delta^ 2/2\). Consequently under the above assumptions \(\| f-B_ nf\|_ p=O(n^{-\alpha})\) is equivalent to \(E_ n(f)_ p=O(n^{-2\alpha})\) where \(E_ n(f)_ p\) is the rate of best polynomial approximation to f in the \(L^ p\)-norm. The paper contains useful lemmas and other interesting results.
0 references
Bernstein polynomials
0 references
rate of best polynomial approximation
0 references