82 papers
The "Stat — Ot" category on Gist.Science focuses on specialized statistical research that falls outside standard classifications, often exploring novel mathematical frameworks for data analysis. While these papers can feel dense to non-experts, they tackle fundamental questions about how we measure uncertainty and interpret complex patterns in the world.
Every new preprint in this field is sourced directly from arXiv, the leading open-access repository for scientific papers. Our team at Gist.Science processes each of these submissions to provide both a clear, plain-language explanation and a detailed technical summary, ensuring that groundbreaking statistical methods are understandable to everyone from students to seasoned researchers.
Below are the latest papers from this category, organized to help you explore the newest advancements in this unique corner of statistics.
This paper proposes a descriptive frequentist framework that reinterprets P-values and S-values not as measures of logical impossibility or direct evidence against a model's truth, but as graded indices of data-model compatibility and refutational information that must be integrated with contextual factors and analyst judgment to determine a model's practical acceptability.
This paper introduces three open-source Shiny web applications—Statistics 101, 201, and 202—that facilitate the teaching of probability distributions, inferential statistics, and simple linear regression by providing interactive, code-free interfaces that display numerical results, visualizations, and mathematical derivations side by side.
This paper employs probabilistic methods and a newly constructed (49, 6, 5) covering design that meets the Schönheim bound to demonstrate that common lottery strategies are mathematically flawed because they restrict number subsets and disproportionately reduce winning chances, ultimately confirming that the house always wins.
This paper advocates for e-values and e-processes as robust measures of statistical evidence, arguing that they synthesize the desirable properties of p-values, likelihood ratios, and Bayes factors while offering unique advantages in handling composite hypotheses, optional stopping, and intuitive wealth-based interpretations.
This paper mathematically demonstrates that the high false positive rates of AI detectors against diverse student populations are an inherent, unavoidable consequence of the statistical overlap between human and AI writing distributions under a composite null hypothesis, proving that no amount of technological improvement can eliminate these structural detection limits.
This paper proposes a Bayesian regime switching model that analyzes longitudinal vital signs and lab data from a large Mayo Clinic dataset to probabilistically detect occult hemorrhage and physiological shock in ICU patients, thereby enabling earlier clinical intervention.
This paper proposes a Bayesian framework that transforms traditional Cornfield-type sensitivity analysis from a threshold-based diagnostic into a probabilistic assessment by modeling confounding strength as a random variable to calculate the posterior probability that unmeasured confounding exceeds the E-value threshold.
This paper proposes an optimization-driven framework for multivariate conformal prediction that jointly learns a predictive model and a novel nonconformity score to generate valid, minimum-volume prediction sets defined by arbitrary norm balls, thereby overcoming the geometric rigidity and computational inefficiency of existing methods.
This paper explores the integration of autocalibration and fairness into the concept of multicalibration for nonlife insurance pricing, proposing practical implementation methods like local regression and bias correction, and validating their relevance through a motor insurance case study.
This paper introduces Kernel Quantum State Estimation (KQSE), a non-parametric framework that robustly reconstructs non-Gaussian continuous-variable quantum states and their trace characteristics from noisy tomographic data without prior knowledge, achieving near-optimal convergence rates.