252 papers
The paper proposes the windowEM algorithm, a stochastic variant of the EM method that partitions data into blocks arranged on a circle to generate a population of estimates via sequential updates and rolling window smoothing, thereby offering convergence guarantees and potential over-fitting prevention.
This paper introduces a quantitative, simulation-based framework that replaces informal sample-size rules with empirically calibrated confidence regions in the skewness–sample-size plane to rigorously determine the applicability of the Central Limit Theorem for both continuous and discrete datasets.
This paper proposes the PLH-GEV, a flexible Power Log Hybrid extension of the Generalized Extreme Value distribution that adaptively corrects survival tail rigidity to significantly improve extreme quantile accuracy in financial and insurance datasets while reverting to the classical model when unnecessary.
This paper derives fundamental properties, including normalization conditions, generating functions, and moment relationships, for the standardized Gram–Charlier Type A distribution and its exponentiated version, while also addressing limitations such as the non-negativity issue of truncated expansions.
This study validates the analytical accuracy of Wistats v3.0 by demonstrating its complete agreement with established statistical software (SPSS) and machine learning libraries (scikit-learn) across diverse analytical scenarios, confirming its reliability as a user-centered platform for automated, publication-ready scientific reporting.
This paper proposes and simulates a two-level monitoring architecture for work-related psychosocial risks that combines an objective, sector-level Organizational Load Index with an individual-level Psychosocial Response Index, demonstrating through Monte Carlo analysis that refined alert criteria can achieve high sensitivity while minimizing false alarms compared to naive thresholding.
This paper introduces the three-parameter Sine Type II Topp-Leone Frechet (STIITLF) distribution, derives its key statistical properties and parameter estimation methods, and demonstrates its superior flexibility and goodness-of-fit compared to existing models through Monte Carlo simulations and applications to real-life lifetime data.
This study develops a comprehensive analytical framework combining ABESS-based variable selection, logistic regression, and Generalized Additive Models to accurately identify and interpret the clinical and behavioral factors driving hospitalization cost overruns for pulmonary infections under DRG payment reforms.
This paper demonstrates that aggregate Poisson predictive log-likelihood is a flawed metric for cross-city crime model comparison due to a scale confound favoring sparser populations, and proposes a superior evaluation framework using nonzero event-weighted likelihood and calibration ratios instead.
This paper presents an exact, differentiable Bayesian inference method for the Hurst index in fractional stochastic volatility models by integrating the Davies–Harte circulant-embedding algorithm into Hamiltonian Monte Carlo, thereby achieving computational efficiency and unbiased posterior estimates that refute the rough volatility hypothesis in favor of long-memory dynamics for the VN30 index, unlike biased Whittle approximations.