Clay weekly context brief for the Statistics category (ISO week 2026-W44).

Clay tracks publications from the Statistics feed list. Below are recent items from this category, each with its source and a short description of what the publication covers when one is available in the source feed.

Recent publications:
1. High-Value Training Data from Noisy, Diverse Sources
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.30421
   Training state-of-the-art models increasingly requires learning from large, heterogeneous datasets whose examples vary in quality, domain relevance, and source of supervision.
2. Phase Transition from Random Features to Fully Trainable Networks
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2610.31663
   Constructing accurate and computationally tractable approximation spaces is a fundamental task in scientific computing.
3. DagFlow AI: Agentic Synthetic Data Generation for Personalized Statistics Learning
   Source: stat.OT (Other Statistics)
   Link: https://arxiv.org/abs/2610.30580
   To provide personalized learning experiences in courses that teach data analysis skills, instructors often search for datasets that appeal to a student's interests, align with specific learning objectives, and are appropriate for the student's skill level.
4. A Physically-Motivated Stochastic Model of Tropical Cyclone Dynamic Wind Fields
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2610.31349
   Tropical cyclones (TCs) pose significant risk to infrastructure and human safety, and the frequency and intensity of large TCs are expected to increase in the future.
5. The Hammock Plot: Where Categorical and Numerical Data Relax Together
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2506.13630
   Effective methods for visualizing data involving multiple variables, including categorical ones, are limited.
6. Estimation of Treatment Effects Under Nonstationarity via the Truncated Policy Gradient Estimator
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2506.05308
   Randomized experiments (or A/B tests) are widely used to evaluate interventions in dynamic systems such as recommendation platforms, marketplaces, and digital health.
7. Flow-based Spectral Kernel Learning for Nonstationary Attention
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.30465
   Kernelized attention provides a useful perspective for understanding self-attention as a similarity-based aggregation mechanism, but existing approaches often rely on fixed or stationary kernels, limiting their ability to adapt the similarity structure to complex token interactions.
8. Robust Power and Sample Size Calculations in Quasi-likelihood Models: Methods and Practice
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2603.00734
   Accurate power and sample size (PSS) calculations are essential for designing studies that use quasi-likelihood (QL) models, which extend generalized linear models (GLMs) to settings where the full distribution of the outcome is not specified.
9. A Bayesian Copula Dirichlet Process GLM for Longitudinal Data
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2610.30467
   We introduce a semiparametric Bayesian generalized linear model (GLM) for longitudinal continuous outcomes that preserves the usual interpretation of the GLM regression structure while leaving the baseline distribution nonparametric.
10. Snapshot plots: displaying summary tables as parallel univariate plots with consistent color highlighting
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2607.28302
   For empirical studies, social and health scientists give background characteristics of their sample and summarize them in the famous "Table 1".
11. A general class of shared gamma frailty models for risk prediction, with application to breast cancer screening adherence
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2610.28548
   We introduce a new class of mixed models for longitudinal or clustered binary data.
12. xVGAE: A Hierarchical Variational Graph Autoencoder for Exchangeable Graphs
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.30815
   Graphs play a crucial role in many applications, ranging from drug discovery to social network modeling and astronomy.
13. Differentially Private Synthetic Data via Generalized Moment Matching
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2610.30196
   We study $(\varepsilon,\delta)$-differentially private synthetic data for arbitrary databases of $n$ records in $[-1,1]^d$ under a specified integral probability metric (IPM).
14. Prediction Uncertainty Controlled Computer Model Calibration
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2610.31163
   Computer-model calibration is usually undertaken in service of prediction, yet neither Bayesian Kennedy-O'Hagan (KOH) calibration nor amortized simulation-based inference is built to control the predictive error of the calibrated simulator.
15. Generalizing Metropolis-Hastings for DAG Sampling
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2610.30291
   A fundamental challenge in Bayesian network structure learning lies in efficiently approximating the posterior distribution of directed acyclic graphs (DAGs).
16. Exact Covariance Structure and Efficient Computation for L-Comoments
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2610.28387
   L-moments are useful summary statistics and so have been widely employed in statistical analysis.
17. Continuous-depth Deep Gaussian Processes
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.30890
   Deep Gaussian processes (DGPs) are appealing Bayesian regression models, but their standard discrete-depth parameterization often becomes harder to optimize as depth grows.
18. Sharp convergence rate of the estimation error of LASSO via the Feature Space Decomposition method
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2610.30622
   In this paper, we use the Feature Space Decomposition (FSD) method to establish an explicit, non-asymptotic, instance-optimal convergence rate for the estimation error of the LASSO in the Gaussian linear regression problem for any signal, any diagonal covariance and any regularization parameter above $\sigma_\xi/\sqrt{N}$ where $\sigma_\xi$ is the noise variance and $N$ the number of data.
19. Detecting adverse high-order drug combinations from individual case safety reports using computational statistics on disproportionality measures
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2504.00646
   Adverse drug reactions linked to the intake of drug combinations are a critical concern in pharmacovigilance, particularly as the controlled environment of clinical trials often lacks the scale and diversity to detect rare events involving multiple medications.
20. Beyond Global Importance: A Heterogeneity Index for Sensitivity Analysis
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2610.31008
   Global sensitivity analysis quantifies how strongly inputs influence a model output, however, input importance often varies across the model space.
21. Interpretable subgroup identification across survival and generalized-linear outcomes
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2610.28379
   A subgroup analysis searching a trial's covariates or many prespecified subgroups often singles out a subgroup for its extremity, interpreted as if prespecified.
22. Random Projections for Spectral Algorithms: Optimal Rates in Sobolev Norms
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.30948
   Random projections (RP) offer an effective method for reducing computational and storage overhead while preserving the geometric structure of data.
23. Finite-Query Coherent Projection of Black-Box Conditional Predictions
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2610.30708
   Local conditional predictions from a black box need not admit a common joint distribution.
24. Penalized Fair Regression for Multiple Groups in Chronic Kidney Disease
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2512.17340
   Fair regression methods have the potential to mitigate societal bias concerns in health care, but there has been little work on penalized fair regression when multiple groups experience such bias.
25. Multichoice Modeling: Exploring Decision Configuration Space under Uncertainty
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2610.31038
   Decision support has become sophisticated at exploring uncertainty, but much less so at representing the architecture of the decision itself.
26. RECAP: Representation-Based Calibration with Amortized Pooling for Frequentist Simulation-Based Inference
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2610.28332
   Simulation-based inference enables uncertainty quantification in broad contexts when a simulation model is available but when a test statistic's sampling distribution is not tractable.
27. Metric Conformal Prediction via the Conditional Expected Local Radius
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.31336
   We introduce the conditional expected local radius (CELR) and develop a new conformal prediction method for non-Euclidean responses in general metric spaces.
28. Functional Inference for the Mean Curve of Intraday Spot Correlation
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2610.31208
   In this paper, we propose estimators for the mean of the daily spot-correlation curves, which are robust to the presence of finite-variation jumps in the price processes.
29. StochIKM: Automated High-Fidelity Reconstruction of Individual Patient Data from Published Kaplan-Meier Curves
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2610.29122
   Individual patient data are the gold standard for secondary analysis of time-to-event evidence, but they are rarely released, and reconstruction from a published Kaplan-Meier figure could help.
30. Computation of Maximum Likelihood Estimates for Multiresponse Generalized Linear Mixed Models with Non-nested, Correlated Random Effects
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/1403.7676
   Estimation of generalized linear mixed models (GLMMs) with non-nested random effects structures requires approximation of high-dimensional integrals.
31. Bayesian nonparametric mixtures of asymmetric Archimedean copulas
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2610.28260
   Archimedean copulas are exchangeable: a single generator parameter controls every pairwise association, and the copula is invariant under permutations of its arguments.

Sources in this brief: stat.AP (Applications); stat.CO (Computation); stat.ME (Methodology); stat.ML (Machine Learning); stat.OT (Other Statistics); stat.TH (Statistics Theory).
Selected 31 of 441 available items for this weekly brief.