Clay weekly context brief for the Statistics category (ISO week 2026-W45).

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. Amortized Localized Posterior Inference for Model Calibration with Application to an Ocean Turbulence Closure
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.37731
   Model discrepancy may prevent a single calibration parameter from reproducing a physical response across the context domain.
2. Outcome-dependent density combination by mixture Bayesian predictive synthesis
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2610.39875
   In the problem of density combination, the decision maker receives multiple density functions and is tasked to learn which density is used most in which region of the outcome space, and to combine them into a single density.
3. Critique of "Use of roster charts in the investigation and prosecution of nurses ..." by John O' Quigley
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2606.29394
   The paper "Use of roster charts in the investigation and prosecution of nurses suspected of inflicting deliberate harm on patients" by Prof.
4. Fisher Geometry of Circular Wrapped Normals: Completion and Curvature
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2610.37985
   We determine the intrinsic Fisher geometry of the two-parameter circular wrapped normal family.
5. Movement synchronization in complex and dynamic team coordination
   Source: stat.OT (Other Statistics)
   Link: https://arxiv.org/abs/2606.21009
   Mutual anticipation of teammates' actions enables efficient coordination that achieves a common goal and high performance.
6. 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.
7. Statistical Analysis of Synthetic Data with Verification: A Linear Regression Framework
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.40254
   Synthetic data are widely used for training modern AI systems.
8. Reference-based multiple imputation for longitudinal bounded discrete outcomes
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2610.39834
   Reference-based multiple imputation has been developed for binary and ordinal responses.
9. Real-time Win Probability and Latent Player Ability via STATS X in Team Sports
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2602.19513
   This study proposes a statistically grounded framework for real-time win probability evaluation and player assessment in score-based team sports, based on minute-by-minute cumulative box-score data.
10. Finite-Sample Optimal Conformal Prediction Sets under Covariate Shift
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2610.37866
   We study conformal prediction under covariate shift with a known covariate density ratio, seeking prediction sets with minimum expected length under the target distribution.
11. Resolving the Mobility-Equity-Emissions Trilemma: A Causal Machine Learning Framework for Dynamic Tolling Policy
   Source: stat.OT (Other Statistics)
   Link: https://arxiv.org/abs/2610.33994
   We propose a causal machine learning framework to integrate dynamic tolling for mobility, equity, and climate objectives, resolving a long-standing policy trilemma.
12. Bayesian Generalized Network Autoregressive Model with Structured Shrinkage and Persistence Priors
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2609.07257
   We propose a Bayesian generalized network autoregressive (BGNAR) model for multivariate time series whose component series are associated with the nodes of a known network.
13. Understanding axial attention in TSFMs through single location regression
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.40188
   Recent advances in Time Series Foundation Models (TSFMs) increasingly rely on axial attention mechanisms, but despite their empirical success, the theoretical properties of axial attention remain largely unexplored.
14. Generalized Posterior Calibration via Infinitesimal-Jackknife Variance Matching
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2610.39684
   Generalized posteriors update a prior through an exponential transformation of empirical loss rather than a conventional likelihood, with a learning rate controlling posterior concentration.
15. Conformal Annotation with Randomized Expert Review: Application to Online Cell-Type Labeling
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2610.40252
   Artificial intelligence (AI) models can accelerate cell-type annotation in digital pathology, but safely scaling their use requires determining, with error control, which model-generated labels can be accepted without expert review.
16. Central Limit Theorems and Network-Free Inference for Heterogeneous Spatial Autoregressions on Sparse Graphons
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2610.40630
   Network dependence is widely regarded as a fundamental obstacle to inference: outcomes linked through a graph are not independent, the graph is rarely observed in full, and models for how dependence propagates are easily misspecified.
17. 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.
18. SENECA: Small-Sample Discrete Entropy Estimation via Self-Consistent Missing Mass
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2605.00668
   Discrete entropy estimation is a classic information theory problem, wherein the average information content of a discrete random variable is estimated from samples alone.
19. Hidden Disparities in Ranking: Category-Aware Fairness via Distributional Stability
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.40182
   We study the problem of fair ranking in recommender systems.
20. Variational Bayes for Variable Selection with Adaptive Sharing across Populations and Responses
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2610.39677
   We propose a variational Bayesian framework for variable selection with multiple responses across heterogeneous populations.
21. TwinPoSI: Statistical Twins for Powerful Post-Selection Inference on Population-Based Regression Coefficients
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2610.39274
   Post-selection inference (PoSI) quantifies uncertainty about regression coefficients after data-driven feature selection.
22. Self-regularization and structural adaptation of NPMLE for Gaussian mixture models in all dimensions
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2610.40566
   We establish structural properties of the nonparametric maximum likelihood estimator (NPMLE) for $d$-dimensional Gaussian mixtures whose means lie in a hypercube.
23. Iterative Identification Closure: Amplifying Causal Identifiability in Linear SEMs
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2604.09309
   The Half-Trek Criterion (HTC) is the primary graphical tool for determining generic identifiability of causal effect coefficients in linear structural equation models (SEMs) with latent confounders.
24. Fisher Contrastive Learning to Mitigate Feature Suppression: A Case Study in Acute Promyelocytic Leukemia Diagnosis
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.40162
   Self-supervised contrastive learning (SSCL) learns representations by contrasting augmented views of the same sample, but it can exhibit feature suppression, where dominant or easy-to-learn factors suppress other class-relevant signals, degrading the performance of representations, particularly in high-dimensional data with heterogeneous features.
25. Nonlinear Sufficient Dimension Reduction for Causal Inference with Continuous Treatments
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2610.39422
   Causal inference with continuous treatments requires estimating nuisance functions to adjust for the potentially high-dimensional confounder, but nonparametric estimation of these functions is severely hampered by the curse of dimensionality.
26. Modular Bayesian Foundation Models: Bayesian time series encoders for OOD detection, routing and forecasting
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2610.39043
   Industrial systems continuously generate time series that vary in sampling frequency, scale, dynamics, and operating regime.
27. Rate-Optimal Spectral Likelihood-Ratio Clustering in Stochastic Block Models
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2610.40531
   Spectral clustering commonly assigns communities by clustering a low-dimensional representation, whereas likelihood refinement can improve an initial assignment using the observed edges.
28. Detecting Model Misspecification in Bayesian Inverse Problems via Variational Gradient Descent
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2512.01667
   Bayesian inference is optimal when the statistical model is well-specified, while outside this setting Bayesian inference can catastrophically fail; accordingly a wealth of post-Bayesian methodologies have been proposed.
29. Reinforcement Learning Under Runtime Confounding: Policy Learning and Off-Policy Evaluation
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.40056
   Reinforcement learning (RL) is increasingly used to train robust, adaptable systems across healthcare, e-commerce, and robotics.
30. Self-normalised change point tests that remain pivotal under non-stationarity
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2610.39404
   Self-normalising test statistics for time series circumvent the lag-specification of long-run variance estimators by employing an inconsistent estimator for studentization.
31. Combining Adaptive Seamless Phase II/III Designs and Group Sequential Analysis for Time-to-Event Endpoints in Oncology Trials
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2610.38301
   Seamless Phase II/III designs are a natural vehicle for the randomized dose optimization called for by the US Food and Drug Administration's Project Optimus initiative.
32. Marginal Composite Likelihood Estimation of Bergomi Models from High-Frequency Option Prices
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2610.40425
   We develop a theory for estimating the kernel parameters of Bergomi-type stochastic volatility models from option-price time series.

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 32 of 420 available items for this weekly brief.