The paper introduces a new estimand for conditional distributional treatment effects that captures how treatments influence the entire outcome distribution, including variance and tail risks, in a covariate-dependent manner. It presents a doubly robust estimator that is minimax optimal locally and uses it to construct a test for global homogeneity of conditional potential outcome distributions. The test accommodates discrepancies beyond the maximum mean discrepancy, guarantees valid type‑1 error, is consistent against fixed alternatives, and includes a computationally efficient, permutation‑free algorithm with exact closed‑form expressions for two natural discrepancies.
By Saksham Jain, Alex Luedtke
arXiv:2507. 12843v3 Announce Type: replace Abstract: Are two distributions close to each other with statistical significance?
By Zhijian Zhou, Liuhua Peng, Xunye Tian, Mingming Gong, Feng Liu
arXiv:2607. 24235v1 Announce Type: cross Abstract: Over the past 20 years, kernel discrepancies have been leveraged as a highly powerful tool for quantifying the disagreement of distributions, with numerous successful applications in two-sample, goodness-of-fit, and independence testing, among others.
By Jose Cribeiro-Ramallo, Florian Kalinke, Zolt\'an Szab\'o
arXiv:2606.00661v2 Announce Type: replace-cross
Abstract: Median-of-means (MoM) is a powerful technique that theoretically enables near sub-Gaussian finite-sample rate for parameter estimation when t...
By Nong Minh Hieu, Antoine Ledent
arXiv:2607. 20119v1 Announce Type: cross Abstract: We introduce the Directional Kernel Mean Difference (DKMD), a signed statistic for univariate distribution comparison that preserves the direction of distributional shifts.
By Shijie Zhong, Jiangfeng Fu
arXiv:2601.13874v3 Announce Type: replace-cross
Abstract: Accurately and efficiently estimating the variance of the Maximum Mean Discrepancy (MMD) remains challenging, particularly for unbalanced sam...
By Shijie Zhong, Yikun Yang, Da Gong, Jiangfeng Fu
The paper investigates Double Machine Learning (DML) estimators under structure‑agnostic (SA) models, which assume the data‑generating law lies within a neighborhood of fixed machine‑learning estimates. It shows that for two of three studied functionals—the quadratic functional in the Gaussian sequence model and the quadratic density integral functional—the DML estimators are asymptotically inadmissible, being dominated by second‑order empirical higher‑order influence function (HOIF) estimators. For the third functional, the expected conditional covariance, both DML and HOIF estimators remain minimax but neither dominates the other.
By Lin Liu, Rajarshi Mukherjee, James M Robins
arXiv:2606.06332v2 Announce Type: replace-cross
Abstract: Asymptotic e-values are emerging as a powerful alternative to asymptotic p-values, particularly in post-hoc inference and multiple testing, w...
By Diego Martinez-Taboada, Ben Chugg, Aaditya Ramdas
arXiv:2504. 19952v2 Announce Type: replace-cross Abstract: We present two general lower bounds for stopping times of sequential tests between arbitrary composite nulls $\mathcal P$ and alternatives $\mathcal Q$.
By Shubhada Agrawal, Ashwin Ram, Aaditya Ramdas
arXiv:2505. 20178v2 Announce Type: replace-cross Abstract: Prediction-Powered Inference (PPI) is a popular strategy for combining gold-standard and possibly noisy pseudo-labels to perform statistical estimation.
By Pranav Mani, Peng Xu, Zachary C. Lipton, Michael Oberst
arXiv:2601. 22784v2 Announce Type: replace-cross Abstract: We introduce a rank-statistic approximation of $f$-divergences that avoids explicit density-ratio estimation by working directly with the distribution of ranks.
By Viktor Stein, Jos\'e Manuel de Frutos
The paper investigates how watermarking affects recursive discrete distribution estimation when synthetic samples are mixed with real data. It establishes minimax lower bounds showing that, as the proportion of real samples approaches zero, adding watermarks cannot improve performance unless the false‑negative detection rate also vanishes. The authors further demonstrate that simple deterministic estimators achieve worst‑case losses close to these bounds and introduce a masking technique that reduces the remaining performance gap to a Jensen gap, suggesting potential for tighter bounds.
By Millen Kanabar, Michael Gastpar