| Simple Universal Bounds for Chebyshev-Type Quadratures (2009) | |||||||||
Abstract | |||||||||
| A Chebyshev-type quadrature for a probability measure sigma is a distribution which is uniform on n points and has the same first k moments as sigma. We give bounds for the smallest possible n required to achieve a certain degree k. In contrast to previous results of this type, our bounds use only simple properties of sigma and are thus applicable in wide generality. In particular, it is shown that whenever sigma has bounded density on a finite interval, n may increase at most exponentially with k. Examples are given illustrating the tightness of our bounds, and applications are given to special local constructions on the sphere and cylinder and to an apparently new result on Gaussian quadrature. We also introduce the concept of random Chebyshev-type quadratures, the case in which nodes are chosen by independent random samples from sigma. The concept is discussed and some preliminary results are proven. These results were recently applied to understand how well can a Poisson process approximate certain continuous distributions. We conclude with a list of open questions.. Comment: 40 pages | |||||||||
Publication details | |||||||||
| |||||||||