We show that under minimal assumptions on a random vector and with high probability, given independent copies of , the coordinate distribution of each vector is dictated by the distribution of the true marginal . Specifically, we show that with high probability, \[\sup_{\theta \in S^{d-1}} \left( \frac{1}{m}\sum_{i=1}^m \left|\langle X_i,\theta \rangle^\sharp - \lambda^\theta_i \right|^2 \right)^{1/2} \leq c \left( \frac{d}{m} \right)^{1/4},\] where and denotes the monotone non-decreasing rearrangement of . Moreover, this estimate is optimal. The proof follows from a sharp estimate on the worst Wasserstein distance between a marginal of and its empirical counterpart, .
View on arXiv