报告时间：2026年6月2日（周二）上午 11:00-12:00
报告地点：91国产 未来校区 未来科创中心412
报告人：叶子馨 博士后，法国萨克雷大学
报告摘要：
We derive general bounds on the rate of convergence with respect to the convex distance in the multivariate stable central limit theorem for all α∈(0,2). Our bounds are derived by combining theory from the recent breakthroughs on Stein's method for α-stable approximation in Wasserstein(-type) distances together with a powerful smoothing inequality and induction. We illustrate the broad applicability of our bounds through five concrete examples. In the case that the summands are distributed as a mixture of two Pareto distributions with tail indices α and α+β for some β>0, we verify that our general bounds are of optimal order with respect to the sample size n for α∈(1,2), of optimal order for α=1 (possibly up to a log factor, depending on the value of β), and of optimal order for α∈(0,1) for sufficiently large β.
报告人简介：
叶子馨，法国萨克雷大学博士后，本科毕业于山东大学和曼彻斯特大学，硕士、博士分别毕业于牛津大学、曼彻斯特大学。主要从事概率论的研究，包括Stein's method，稳定分布，随机微分方程参数的估计等，在Stud. Appl. Math.，J. Math. Anal. Appl.等期刊发表多篇文章。
邀请人：董自康