English

Expected $L_2-$discrepancy bound for a class of new stratified sampling models

Statistics Theory 2022-04-20 v1 Probability Statistics Theory

Abstract

We introduce a class of convex equivolume partitions. Expected L2L_2-discrepancy are discussed under these partitions. There are two main results. First, under this kind of partitions, we generate random point sets with smaller expected L2L_2-discrepancy than classical jittered sampling for the same sampling number. Second, an explicit expected L2L_2-discrepancy upper bound under this kind of partitions is also given. Further, among these new partitions, there is optimal expected L2L_2-discrepancy upper bound.

Keywords

Cite

@article{arxiv.2204.08752,
  title  = {Expected $L_2-$discrepancy bound for a class of new stratified sampling models},
  author = {Jun Xian and Xiaoda Xu},
  journal= {arXiv preprint arXiv:2204.08752},
  year   = {2022}
}

Comments

arXiv admin note: substantial text overlap with arXiv:2203.01288

R2 v1 2026-06-24T10:51:52.533Z