stat papers | Gist.Science

Agreement is not corroboration: bounding the independence of cultural-heritage authority links

This study demonstrates that agreement between Getty ULAN and Library of Congress authority records in Wikidata does not guarantee independent corroboration, as a significant portion of links are propagated or untraceable, resulting in an independence rate that remains statistically indeterminate and highlights the critical need to explicitly model provenance dependence in linked-data workflows.

Emin Talip Demirkiran2026-09-17

📊 statistics

Extending Deep Cox Survival Models with Case-Cohort risk set Sampling

This paper introduces and evaluates the first integration of case-cohort risk-set sampling into deep Cox proportional hazards neural networks, demonstrating that while nested case-control sampling consistently approximates full-risk-set performance, mini-batch training enables case-cohort sampling to achieve comparable accuracy with substantial computational savings.

Justine Nasejje, Michael P. Falk, Marvin N. Wright, Wende Clarence Safari, Pierre-Philippe Dechant, Raeesa Manjoo Docrat (…)2026-09-17

📊 statistics

Comparative Evaluation of Dynamic Bayesian Hierarchical Models and Machine Learning Estimators for Small Area Employment Estimation in Data-Poor Settings

This study demonstrates that while Dynamic Bayesian Hierarchical Models outperform machine learning estimators in accuracy and uncertainty quantification for small area employment estimation under severe data scarcity, machine learning methods become competitive as sample sizes and auxiliary data quality improve, offering practical guidance for method selection in data-poor settings.

Albert Schulle2026-09-16

📄 social science

Implementing Flipped Classroom in a Biomathematics Course Through Design-Based Research

This Design-Based Research study on implementing a Flipped Classroom model in a biomathematics course found that while students achieved exceptionally high academic performance, their self-assessed teamwork attributes generally declined, likely due to a "response shift" caused by the increased cognitive load of mastering complex software tools.

Gergő Sipka, Péter Gaszler, Mónika Ágnes Szütsné Tóth, Beáta Bugyi2026-09-15

📊 statistics

How small a difference can conductometry establish? An uncertainty budget, resolution curves and the arithmetic cost of four design faults

This paper constructs a GUM-compliant uncertainty budget and resolution curves to demonstrate that common experimental design faults, particularly uncontrolled temperature and statistical missteps, drastically inflate false-positive rates in conductometric studies, ultimately proving that the author's own previously reported data fails to meet the rigorous detection limits required to claim altered physicochemical properties.

Vadim Kovalev2026-09-15

📊 statistics

Public acceptance of artificial intelligence in health care: a cumulative hierarchy in a 16-item instrument, German–Italian measurement equivalence, and a five-item short form

This study demonstrates that public acceptance of artificial intelligence in healthcare follows a cumulative hierarchy ordered by clinical consequence, remains invariant across German- and Italian-speaking populations, and can be effectively captured by a five-item short form despite the lack of a uniquely determined item set.

Dietmar Ausserhofer, Christian J. Wiedermann, Verena Barbieri, Nicoletta Colletti, Stefano Lombardo, Timon Gärtner, Giul (…)2026-09-15

📊 statistics

Predictive Forecasting of Care Load

This study develops and deploys an end-to-end forecasting system using engineered features and machine learning models to predict near-term care loads for Unaccompanied Alien Children in the U.S. HHS program, while transparently documenting a performance discrepancy between a persistence baseline and the final Gradient Boosting model to support staffing and capacity-planning decisions.

Sreethi Kamuju, M.Ramchander2026-09-15

⚡ electrical engineering

Why Variance Metrics Cannot Validate ComBat Harmonization: Evidence from Three-Vendor Mammography Radiomics

This study demonstrates that conventional variance-based metrics provide a tautological and misleading validation of ComBat harmonization, as they fail to detect persistent non-linear vendor-specific patterns and biological signal loss, thereby necessitating the use of independent non-linear separability tests to truly assess cross-vendor comparability.

XU ZHANG, YUNJIANG LI, RUI FENG, QINQIN AI, YUHAI XIE, WENBIN NI2026-09-15

📊 statistics

Spectral Resonance Decomposition: Causal-Invariant Adaptive Basis Modeling with Calibrated Uncertainty

This paper introduces AstralModel, a novel predictive framework utilizing Spectral Resonance Decomposition to synthesize causal-invariant, interpretable non-linear basis functions that match the accuracy of tree-based ensembles while providing rigorous uncertainty quantification and transparency.

Aditya Pandey2026-09-15

📊 statistics

Lévy-Regularized Bayesian Neural Networks for Fair Reserve Estimation under Solvency II

This paper proposes a Lévy-regularized Bayesian neural network framework that integrates measure-theoretic valuation, tempered stable noise injection, and causal fairness adjustments to significantly improve the accuracy and equity of life insurance reserve estimation under Solvency II, particularly during extreme demographic shocks like pandemics.

Antonio Nazzaro2026-09-15

← Previous Next →