Optimal recovery of the derivative of periodic analytic functions from Hardy classes (Q1284499): Difference between revisions

Let \(X\) and \(Y\) be linear spaces, \(L\) a linear functional on \(X\), and \(I : X\to Y\) a linear operator. Suppose that \(W\subset X\). The quantity \(e(L,W,I)=\inf_{F}\sup_{x\in W} | Lx-F(Ix)| \), where \(F\) is any functional on \(Y\) (not necessarily linear or continuous) is called the intrinsic error of the problem of the optimal recovery of \(Lx\), \(x\in W\), on the basis of the information \(Ix\). In this paper the author considers this problem in the case where \(X\) is the space \(H_{p,\beta}\), \(1\leq p\leq \infty\), of all \(2\pi\)-periodic functions \(f\), which are analytic in \(S_{\beta}=\{z\in {\mathbb C}: | \Im z| <\beta\}\) and satisfy \[ \| f\| =\sup_{0\leq\eta <\beta} ({1\over 4\pi}\int_0^{2\pi} (| f(t+i\eta)| ^p+| f(t-i\eta)| ^p)dt)^{1/p} <\infty, \] \(W=\{f\in H_{p,\beta}:\| f\| \leq 1\}\), \(Y={\mathbb C^n}\), \(Lf =f(\xi)\), \(\xi \in S_{\beta}\), \(If=(f(x_1),\dots,f(x_n))\), where \(x_j\) are distinct points from \({\mathbf T} = [0,2\pi)\). He constructs in terms of elliptic functions an optimal method of recovery and calculates the intrinsic error.