On Low-Dimensional Projections of High-Dimensional Distributions

TL;DR

This paper characterizes when high-dimensional distributions, when projected onto lower dimensions, resemble Gaussian mixtures, establishing necessary and sufficient conditions and analyzing empirical process behavior under such projections.

Contribution

It provides a complete characterization of the Diaconis-Freedman effect, including necessary and sufficient conditions, and studies empirical process behavior under random projections.

Findings

Conditions for the Diaconis-Freedman effect are both necessary and sufficient.

Empirical process behavior under random projections is analyzed.

The effect persists under increasing dimension with specific asymptotic conditions.

Abstract

Let $P$ be a probability distribution on $q$-dimensional space. The so-called Diaconis-Freedman effect means that for a fixed dimension $d << q$, most $d$-dimensional projections of $P$ look like a scale mixture of spherically symmetric Gaussian distributions. The present paper provides necessary and sufficient conditions for this phenomenon in a suitable asymptotic framework with increasing dimension $q$. It turns out, that the conditions formulated by Diaconis and Freedman (1984) are not only sufficient but necessary as well. Moreover, letting $\hat{P}$ be the empirical distribution of $n$ independent random vectors with distribution $P$, we investigate the behavior of the empirical process $\sqrt{n}(\hat{P} - P)$ under random projections, conditional on $\hat{P}$.