85 papers
This paper introduces Efficient Recursive Data Snooping (ERDS), an exact recursive reformulation of iterative data snooping that significantly reduces computational costs for correlated observations by replacing repeated model re-decompositions with closed-form updates, achieving a 96.5% runtime reduction in GNSS data processing while maintaining identical detection results to classical methods.
This paper establishes unified exponential-type concentration bounds for ordered spacings across a diverse set of distributions—including uniform, exponential, normal, hypergeometric, Beta, and Gamma—by leveraging Chernoff–Hoeffding techniques and probability integral transforms to analyze how tail geometry influences bound sharpness.
Using a supervised machine learning pipeline on a synthetic dataset calibrated to US national surveys, this study demonstrates that structural socioeconomic factors, particularly education and occupational class, are the dominant predictors of health vulnerability—explaining 82% of feature importance and supporting fundamental cause theory over behavioral explanations.
This paper introduces HS3, a new implementation-agnostic, human-readable, and extensible serialization standard for statistical models in high-energy physics designed to overcome the limitations of existing formats by enabling machine-readable interoperability, long-term preservation, and FAIR data principles across diverse software frameworks.
This study demonstrates that a hybrid approach combining Isolation Forest anomaly detection with linear mixed-effects modeling can synthesize reliable, optimized pregnancy conception dates from conflicting routine clinical data in low-resource settings, effectively overcoming the lack of early ultrasound and inconsistent record-keeping.
This study proposes a covariance-aware compromise allocation framework for multivariate stratified sampling under nonlinear budget and time constraints, demonstrating through numerical analysis that incorporating covariance interactions yields optimal sample allocations and enhanced precision compared to traditional methods.
This study employs a two-state Markov-switching model with bootstrap uncertainty quantification on Kenya's 2000–2023 maternal mortality data to reveal distinct high and low mortality regimes driven by structural health system changes, demonstrating that the country's maternal health trajectory is characterized by persistent regime shifts rather than a continuous linear trend.
This study proposes a Dynamic Huber-Weighted Adaptive EWMA Max-M statistic that outperforms traditional models in simultaneously detecting both minor and major multivariate process shifts by dynamically adjusting smoothing parameters based on deviation magnitude, as validated through simulations and industrial cement clinker data.
This paper proposes a novel auxiliary-enhanced survey calibration framework for reject inference that avoids the performance degradation of traditional imputation methods by reweighting accepted applicants to match population totals, thereby proving consistent and asymptotically normal while significantly improving population risk calibration and maintaining discrimination power under model misspecification.
This study proposes and validates an interpretable statistical learning framework that integrates penalized regression, bootstrap stability selection, and SHAP-based explainable machine learning to accurately predict student stress, identifying key predictors such as examination pressure and sleep hours while ensuring model transparency and reproducibility.