Clay weekly context brief for the Statistics category (ISO week 2026-W41).

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. Risk-Calibrated Balancing for High-Dimensional Causal Extrapolation
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2609.34482
   In observational causal inference, covariate balancing is widely used to reduce source-target covariate shift, but under weak overlap in high dimensions, stronger balance can induce concentrated weights and increase variance.
2. [DeepGOF] Where Does a Logistic Risk Model Fail? An Audited Neural Goodness-of-Fit Test for Model Development and External Validation
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2609.29575
   Goodness-of-fit tests for logistic regression are routine in clinical risk modelling, yet the classical tests say whether a model misfits, not where.
3. Ready for the Clinic? A Survey of Open-Source Software for Response-Adaptive Randomization in Clinical Trials
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2609.35939
   Response-adaptive randomization (RAR) modifies treatment allocation probabilities during a clinical trial as response/outcome data accumulate, with the aim of improving patient benefit, statistical efficiency, or both.
4. Estimating axial symmetry using random projections
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2512.21417
   We study the directions of axial symmetry of multivariate distributions and relate the measure or cardinality of this set to spherical symmetry.
5. When May a Bandit Leave Its Anchor? E-Process-Authorized Thompson Sampling under Non-stationarity
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.03646
   Stationarity rewards memory, but after a change the same history can mislead.
6. Detection coherence of tests
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2608.11177
   There exist tests calibrated under a null narrower than the one implied by their test statistic -- the detection-null set.
7. A Semiparametric Nonlinear Mixed Effects Model with Penalized Splines Using Automatic Differentiation
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2603.11728
   We present an estimation procedure for nonlinear mixed-effects models in which the population trajectory is represented by penalized splines and adapted to individuals via subject-specific transformation parameters.
8. A Framework to Quantify the Probability of Future Cyber Loss Events
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2609.21717
   Cybersecurity risk quantification remains challenging due to limited operational data and difficulties in quantifying Loss Event Frequency (LEF).
9. Level sets and maximum likelihood estimation for the Ising model
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2511.20925
   Bogdan et al.
10. Broken scale symmetries in undercomplete linear autoencoders
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.03640
   Neural network loss landscapes have many symmetries, which are preserved by gradient flow but broken by finite-stepsize stochastic gradient descent (SGD).
11. Uncertainty-Aware Ideal Point Estimation via Variational EM with P\'{o}lya-Gamma identity
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2605.19591
   Roll-call data analysis aims to estimate legislators' ideal points and quantify the associated uncertainty.
12. Optimal Quantum Speedups for Repeatedly Nested Expectation Estimation
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2602.08120
   We study the estimation of repeatedly nested expectations (RNEs) with a constant horizon (number of nestings) using quantum computing.
13. DOMIC: Provably Calibrated Detection of Dependence-Structure Change Points via Density-Operator Mutual Information
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2609.02787
   Dependence can change between variable blocks without changing correlation; serial dependence complicates calibration.
14. High-Dimensional Asymptotics of Differentially Private PCA
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2511.07270
   In differential privacy, random noise is introduced to privatize summary statistics of a sensitive dataset before releasing them.
15. Below what training size do deep tabular generators stop beating trivial baselines? A preregistered benchmark on a size ladder of clinical and standard datasets
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.03500
   Deep tabular generative models are benchmarked on datasets with tens of thousands of rows; clinical datasets have hundreds.
16. Coarse-to-fine spatial GLMM for scalable prediction and multiscale analysis
   Source: stat.ME (Methodology)
   Link: https://arxiv.org/abs/2605.01157
   We develop a coarse-to-fine generalized linear mixed model (CF-GLMM) for scalable spatial prediction of exponential-family responses.
17. A fast non-reversible sampler for Bayesian mixture models
   Source: stat.CO (Computation)
   Link: https://arxiv.org/abs/2510.03226
   Mixtures models are a cornerstone of Bayesian modelling, and it is well-known that sampling from the resulting posterior distribution can be a hard task.
18. [EDGE] A grouped calibration test for logistic regression that tolerates a few corrupted records
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2608.20511
   The Hosmer-Lemeshow calibration test groups patients by predicted risk for a valid reference and loses power that more groups cannot recover.
19. Factorization by extremal privacy mechanisms: new insights into efficiency
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2507.21769
   We study the problem of efficiency under $\alpha$ local differential privacy ($\alpha$ LDP) in both discrete and continuous settings.
20. Muon Learns Facts Better: Understanding the Role of Spectral Orthogonalization
   Source: stat.ML (Machine Learning)
   Link: https://arxiv.org/abs/2610.02798
   The Muon optimizer applies spectral orthogonalization to matrix-valued updates and has shown strong performance in large-scale neural network training, yet the mechanisms of this transformation in feature learning remain poorly understood.

Sources in this brief: stat.AP (Applications); stat.CO (Computation); stat.ME (Methodology); stat.ML (Machine Learning); stat.TH (Statistics Theory).
Selected 20 of 76 available items for this weekly brief.