stat papers | Gist.Science

Domain-Adaptive Cuckoo Search for Structured Statistical Optimization: Applications to Principal Curves, Single-Cell Trajectory Inference, and Exact Experimental Design

This paper proposes three domain-adaptive variants of the Cuckoo Search algorithm tailored for structured statistical optimization problems in principal curve estimation, single-cell trajectory inference, and exact experimental design, demonstrating their superior performance over standard metaheuristics in multimodal and mixed-integer settings while highlighting the trade-off between accuracy and computational cost.

Elvis Han Cui, Heather Xihe Yu, Weng Kee Wong2026-08-26

📊 statistics

The Complementary Bell-Kumaraswamy-G Family of Distributions: A Flexible Class for Modeling Lifetime and Reliability Data

This paper introduces the novel Complementary Bell-Kumaraswamy-G (CBell-Kw-G) family of distributions, a flexible statistical framework for modeling complex lifetime and reliability data that is rigorously analyzed through theoretical derivation, maximum likelihood estimation, and validation on real cancer survival datasets.

Seyed Jamal Khorashadizadeh, Fatemeh Yousefzadeh2026-08-25

📊 statistics

Mathematical modeling of malaria transmission dynamics in children under five years of age facing innovative interventions in the Democratic Republic of the Congo

This study utilizes a deterministic SEIR-SEI mathematical model to demonstrate that interrupting malaria transmission among children under five in the Democratic Republic of the Congo is mathematically feasible only through a synergistic combination of innovative interventions achieving 80% coverage, as individual strategies or current coverage levels remain insufficient to reduce the basic reproduction number below unity.

Packy Mukanya Mbayo, André Ngombe Kaseba, Jonathan Mubenga Mubenga, Kennedy Lobukulu, Wilfried Nlandu Tembele, Michel Ka (…)2026-08-25

📊 statistics

Evaluating targeted anti-fraud communication in rural China: a cluster-randomised intervention simulation

This methodological simulation study evaluates the statistical power and design behavior of a proposed cluster-randomised field experiment in rural China comparing targeted anti-fraud communication with non-tailored education, demonstrating that the design can effectively detect primary effects while confirming that the reported results reflect simulation properties rather than real-world intervention efficacy.

Yichuan Zhou, Hongying Hu2026-08-25

📊 statistics

The declared winner of a clinical prediction model comparison is often decided by the random split: a simulation study

This simulation study demonstrates that at the small effect sizes typical of clinical prediction model comparisons, the declared "winner" is frequently determined by random data splitting rather than true model superiority, leading to inflated performance margins and significant disagreement between analysts using the same data.

Ananya Sharma2026-08-25

📊 statistics

Optimal Control of Poliovirus Transmission Through Vaccination and Booster Immunization: A Mathematical Modeling Perspective

This study develops an optimal control framework for an extended SEIVBPR model to determine the most cost-effective strategy of combining routine vaccination, booster campaigns, and contact-reduction measures, demonstrating that such interventions can eradicate poliovirus transmission and reduce permanent paralysis by over 80% within 25 weeks.

Timothy Shamaki2026-08-25

📊 statistics

Does Open Peer Review Correlate With Better Altmetric Attention Performance?An Empirical Analysis Based on Nature Communications

Based on an empirical analysis of 35,852 papers from *Nature Communications* using propensity score matching, this study demonstrates that Open Peer Review significantly correlates with higher altmetric attention scores, suggesting that transparent review processes enhance the social impact and public dissemination of scientific research.

Xiaotian WEI, Zhengyi GUAN, Chenyang WANG, Jianxun CHU2026-08-25

📊 statistics

Singular model selection for trustworthy label-free classifier evaluation

This paper demonstrates that label-free classifier evaluation is a singular statistical problem where classical model selection criteria fail due to degenerate Fisher information, and proposes using singular and widely applicable Bayesian information criteria to accurately recover latent structures and ensure trustworthy performance estimates without ground-truth labels.

Vincent Looten2026-08-24

📊 statistics

Integrating CRITIC and Weighted K-Means for Humanitarian Relief Center Location Planning: Evidence from Gaziantep

This study proposes a hybrid decision-support framework that integrates the CRITIC method with a weighted k-means algorithm to optimize the strategic placement of humanitarian relief centers in park areas within Gaziantep, ensuring needs-based and objective planning for post-disaster scenarios.

Zeynep Yüksel, İbrahim Miraç Eligüzel, Suleyman Mete2026-08-24

📊 statistics

Impact of Missing Data Imputation on The Structure of Real-World Clinical Datasets: A Comparison of Median, K-Nearest Neighbours, and MICE Approaches

This study demonstrates that k-nearest neighbours (KNN) imputation outperforms both median substitution and multiple imputation by chained equations (MICE) in preserving the distributional fidelity and multivariate structural integrity of real-world clinical datasets across varying levels of missingness.

Karol Aguiar Quevedo, Alilis Fontana Bellorín, Carlos Jordá Jordá Aragón, Cristóbal Aguilar-Gallardo2026-08-24

← Previous Next →