Sharp inequalities for approximations of convolution classes on the real line as the limit case of inequalities for periodic convolutions (Q2360283)

This article is devoted to the problem of sharp estimates for approximations of convolution classes by entire functions. The history of this problem is presented in the introduction; more results can be found in the books (1) [\textit{N. I. Akhiezer}, Lectures on approximation theory. (Russian) Moscow: Nauka (1965; Zbl 0133.01602)]; (2) [\textit{A. Pinkus}, n-widths in approximation theory. Berlin etc.: Springer-Verlag (1985; Zbl 0551.41001)]; (3) [\textit{A. I. Stepanets}, Methods of approximation theory. Utrecht: VSP (2005; Zbl 1082.42001)]. The author establishes sharp estimates for the best approximations of convolution classes by entire functions of exponential type. To this aim he proposes a new method for testing Nikolskii-type conditions, based on periodization with subsequent passage to the limit (a trick used by the author and \textit{A. V. Gladkaya} [J. Math. Sci., New York 217, No. 1, 3--22 (2016; Zbl 1355.41007); translation from Zap. Nauchn. Semin. POMI 440, 8--35 (2015)] in order to prove sharp inequalities for approximations of classes of differentiable functions by splines). Examples are given, providing sharp estimates for approximations of convolution classes with variation diminishing kernels and generalized Bernoulli and Poisson kernels.