Projections of probability distributions: A measure-theoretic Dvoretzky theorem
Abstract
Many authors have studied the phenomenon of typically Gaussian marginals of high-dimensional random vectors; e.g., for a probability measure on , under mild conditions, most one-dimensional marginals are approximately Gaussian if is large. In earlier work, the author used entropy techniques and Stein's method to show that this phenomenon persists in the bounded-Lipschitz distance for -dimensional marginals of -dimensional distributions, if . In this paper, a somewhat different approach is used to show that the phenomenon persists if , and that this estimate is best possible.
Keywords
Cite
@article{arxiv.1102.3438,
title = {Projections of probability distributions: A measure-theoretic Dvoretzky theorem},
author = {Elizabeth Meckes},
journal= {arXiv preprint arXiv:1102.3438},
year = {2011}
}
Comments
Minor errors corrected resulting in cosmetic changes to statements of some theorems. Assumptions on the underlying distribution have been clarified; a small amount of additional discussion added