Cramér-type Large deviation and non-uniform central limit theorems in high dimensions
share
Central limit theorems (CLTs) for high-dimensional random vectors with dimension possibly growing with the sample size have received a lot of attention in the recent times. Chernozhukov et al., (2017) proved a Berry--Esseen type result for high-dimensional averages for the class of hyperrectangles and they proved that the rate of convergence can be upper bounded by n^-1/6 upto a polynomial factor of p (where n represents the sample size and p denotes the dimension). In the classical literature on central limit theorem, various non-uniform extensions of the Berry--Esseen bound are available. Similar extensions, however, have not appeared in the context of high-dimensional CLT. This is the main focus of our paper. Based on the classical large deviation and non-uniform CLT results for random variables in a Banach space by Bentkus, Rackauskas, and Paulauskas, we prove three non-uniform variants of high-dimensional CLT. In addition, we prove a dimension-free anti-concentration inequality for the absolute supremum of a Gaussian process on a compact metric space.
READ FULL TEXT