stat papers | Gist.Science

The Smallest Standard Error Is the Wrong One: Selection Among Multiple Imputations Invalidates Inference

This paper demonstrates that selecting the multiply imputed analysis with the smallest standard error systematically invalidates statistical inference by discarding essential between-imputation variance and artificially depressing within-imputation variance, leading to severely under-covered confidence intervals and biased estimates, whereas Rubin's pooling rules correctly preserve calibration.

Nyiko Muhluri Khoza¹, Bhekisipho Twala², Jean-Claude Malela-Majika³2026-10-21

📊 statistics

Manual double entry of study data: potential to avoid 4 errors per 1000 entries

This study of 6,730 manually double-entered questionnaires reveals that the process prevents approximately four errors per 1,000 entries, with error rates significantly higher in complex questions and among less experienced data entry clerks.

Kastytis Šmigelskas, Monika Žemaitaitytė, Jūratė Koreivaitė, Vladas Golambiauskas2026-10-21

📊 statistics

Collaborative Mass Matrix Adaptation for Multi-Chain Hamiltonian Monte Carlo

The paper proposes Collaborative NUTS (C-NUTS), a method that pools gradient and position information across multiple parallel chains during warmup to construct a more stable mass matrix estimate, significantly improving sampling efficiency and effective sample size compared to standard single-chain adaptation, particularly in moderate-dimensional settings.

Fen Chen, Pinghua Yang, Zhaohui Chen, Jing Li, Xin Lv2026-10-21

📊 statistics

Early-Career Research Funding and Cumulative Advantage in Scientific Careers: A Mathematical Model and Empirical Analysis

By combining a mathematical model of mutual reinforcement with empirical analysis of 40,000 Japanese researchers, this study demonstrates that receiving early-career research funding creates a statistically significant and widening performance gap over decades, confirming that the "Matthew effect" drives long-term stratification in scientific careers even when controlling for initial talent.

Yuhei Yamaguchi2026-10-21

📊 statistics

Complex Probabilistic Neutrosophic Sets for Multi-Criteria Decision-Making: New Aggregation Operators and Comparative Analysis

This paper introduces the novel complex probabilistic neutrosophic (CPN) set to better capture expert opinion acceptance levels in uncertain real-life scenarios, proposes new weighted averaging and geometric aggregation operators, and validates a corresponding multi-criteria decision-making framework through a car selection case study and comparative analysis.

Faisal Al-Sharqi, Ashraf Al-Quran2026-10-21

📊 statistics

Computational Growth Modeling: Assessing Longitudinal Change with Joint Models

This paper introduces and validates "computational growth modeling," a joint modeling approach that integrates hierarchical computational models with longitudinal data to more accurately estimate individual changes in behavioral processes over time compared to traditional two-stage methods, thereby offering improved insights into the etiology of psychopathology.

Jeremy Haynes, Nathaniel Haines, Holly Sullivan-Toole, Thomas Olino2026-10-21

📊 statistics

Assessing the accuracy of census-independent small area modelled population datasets

This study validates the accuracy of census-independent modelled population estimates by comparing them with Zambia's 2022 census data, finding strong correlations at national and sub-national levels while identifying that refining residential settlement stratification is crucial to reducing systematic overestimation.

Heather R. Chamberlain, Claire A. Dooley, Frank Kakungu, Andrew J. Tatem2026-10-21

📊 statistics

Scalable Bayesian maximum entropy for large environmental data sets through Vecchia graphical inference

This paper introduces VISTA-BME, a scalable method that combines Vecchia graphical inference with Gaussian expectation and mixture propagation to overcome the computational limitations of conventional Bayesian maximum entropy, enabling efficient processing of large environmental datasets with both exact and soft observations while maintaining high accuracy and significantly reducing memory and time costs.

Ying-Fan Lin, Hwa-Lung Yu2026-10-20

📊 statistics

Source information limits lower-risk claims from synthetic validation cohorts

This study demonstrates that while enlarging synthetic validation cohorts can improve conditional precision, it often increases false certification rates for lower-risk claims unless source refitting is applied to mitigate excess errors, though such corrections come with a power cost and do not fully eliminate biases under fixed misspecification.

Tianyi Yu2026-10-20

📊 statistics

Exact Likelihood-Coin Poisson Sampling for Bayesian Inverse Problems with Sharp Complexity Bounds

This paper introduces an exact Bayesian computational framework for inverse problems that utilizes a Bernstein-Poisson construction to convert Bernoulli forward events into likelihood coins, enabling model-independent posterior sampling with sharp complexity bounds and a continuum realization for function-valued elliptic problems.

Zhiliang Deng, Xiaomei Yang2026-10-20

← Previous Next →