European Journal of Statistics

Parametric Versus Non-Parametric Statistics: A Case Study in Traditional and Alternative Medicine Research with the Development of SDA4AMR Web Application

Jularat Chumnaul — 2026-03-30

This study investigated the performance of parametric and non-parametric tests, specifically, the one-sample t-test and the Wilcoxon Signed-Rank (WSR) test, through simulation-based evaluations of type I error rates and statistical power across varying sample sizes and effect sizes. Complementing the simulation study, we conducted a systematic review of empirical articles published in two journals in traditional and alternative medicine to assess the usage of statistical methods and the reporting of assumption checks. Simulation results revealed that the t-test generally maintained acceptable type I error rates under Bradley’s criterion, especially when sample sizes were 20 or greater. In contrast, the WSR test frequently exhibited inflated error rates, particularly with larger samples. In terms of power, the t-test consistently outperformed the WSR test, though both achieved satisfactory power (≥ 0.8) under appropriate conditions. These findings underscore the importance of context-specific test selection, striking a balance between robustness and statistical sensitivity. Furthermore, the journal review revealed that, although a majority of articles employed inferential statistics, assumption checking was rarely reported, even for commonly used methods such as t-tests, ANOVA, and chi-square tests. This lack of transparency raises concerns about the validity of statistical conclusions drawn in the literature. Therefore, to support better statistical practice, the Smart Data Analysis Web Application for Alternative Medicine Research (SDA4AMR) was developed. This tool facilitates the selection of appropriate tests, automatic assumption checking, and interpretable outputs. By enhancing accessibility and encouraging methodological rigor, SDA4AMR aims to improve research quality and reproducibility in the field of traditional and alternative medicine.

Modeling Maternal Blood Loss Using the Exponentiated Kumaraswamy–Inverse Lomax Distribution: Applications to Diverse Real-Life Data

Benson Ade Eniola Afere — 2026-03-30

This study introduces the Exponentiated Kumaraswamy–Inverse Lomax (EK–IL) distribution as a flexible and robust statistical model for analyzing maternal blood loss during delivery. The proposed distribution effectively accommodates skewness and heavy-tailed behavior, which are common characteristics of clinical data. Model parameters are estimated using the maximum likelihood method, and the performance of the EK–IL distribution is evaluated through goodness-of-fit measures and information criteria. Comparative analyses demonstrate that the proposed model outperforms several well-known competing distributions. Further validation using four additional real datasets confirms the adaptability and robustness of the EK–IL distribution. The results suggest that the EK–IL model provides a powerful framework for medical data analysis and broader applications in applied statistics.

Adaptive Forecasting of Epidemiological Time Series with Data-Driven Structural Break Detection: A Comparative Study of Enhanced ARIMA, GAM, and Piecewise Models

Mohamed Alahiane — 2026-03-30

Accurately forecasting epidemic dynamics is a central challenge in statistical epidemiology, particularly when structural breaks induced by policy interventions or behavioral shifts violate the stationarity assumptions of standard forecasting models. This study introduces a unified framework that augments three widely used model classes—Autoregressive Integrated Moving Average (ARIMA), Generalized Additive Models (GAM), and Piecewise Regression—by embedding an endogenous, data-driven change-point detection mechanism based on volatility shifts. We contrast the performance of conventional baseline models with their adaptive “Change-Point” (CP) counterparts, which explicitly incorporate statistically significant volatility-driven regime changes. Using daily COVID-19 incidence data, we conduct a rigorous comparative evaluation of these approaches. We expand the evaluation metrics to include Symmetric Mean Absolute Percentage Error (SMAPE) and Theil’s U statistic to ensure robustness. Our findings show that systematically accounting for structural breaks consistently enhances predictive accuracy across all model families. A notable bias–variance trade-off emerges: while the flexible Change-Point GAM (CP-GAM) attains the best in-sample fit (Adjusted R² = 0.955), the more parsimonious CP-Piecewise model delivers superior out-of-sample forecasts, achieving the lowest Root Mean Square Error (RMSE) and favorable information criteria. Overall, this work offers a statistically principled methodology for modeling nonstationary epidemiological time series and provides reliable forecasting tools to support evidence-based public health decision-making.

On a Discrete New Generalized Pareto-Based Regression Model for Count Data

Jiji Jose — 2026-03-30

Count data often exhibit overdispersion, heavy tails, and decreasing failure rates, which limit the applicability of classical Poisson regression models. In this paper, we develop a regression model based on the discrete new generalized Pareto (DNGP) distribution to better capture these features. The proposed model incorporates covariate effects through suitable link function and parameters are estimated using maximum likelihood estimation methods. Simulation studies, model comparison and real data applications demonstrate that the DNGP regression model provides a flexible and effective alternative for analyzing complex count data.

A Four-Parameter Generalization of the Chen Distribution Family

Scott Smith — 2026-02-25

As an addition to the known continuous distribution families, a four-parameter model of significant flexibility is introduced for the purpose of modeling positive random variables. After the Weibull-Chen {Weibull} Type I model is introduced, some standard numerical properties are explored, including those involved in maximum likelihood estimation. We compare the performance of the model with similar models with application to two real data sets and demonstrate the improved fit.

Enhanced Possibilistic Fuzzy C-Means Clustering Stunting Prevalence in Indonesia

Alivia F. Zahro — 2026-02-24

Stunting in toddlers is a significant public health problem in Indonesia due to its potential to inhibit child development and cause long-term adverse effects. Clustering the prevalence of stunting provides valuable insights for designing effective prevention policies. This study employs the Possibilistic Fuzzy C-Means (PFCM) method, validated using the Modified Partition Coefficient (MPC) index, to cluster stunting prevalence in Indonesia. The PFCM method integrates Fuzzy C-Means (FCM) and Possibilistic C-Means (PCM), balancing membership degrees with probabilistic measures. The primary advantages of this method are its capability to handle data with uncertain membership degrees, robustness against noise, and flexibility in defining probabilistic membership values. The results obtained show that clusters with high stunting prevalence are dominated by nine provinces, namely Aceh, Jambi, Bengkulu, Bangka Islands, Central Kalimantan, Central Sulawesi, Gorontalo, West Papua, and Papua. The MPC validity score of 0.704 confirms the effectiveness of the PFCM method in categorizing stunting prevalence well, making it a robust tool to support policymaking in stunting prevention efforts.

A Note on Local Maxima in Quadratic Transmuted Distributions Likelihoods

Edoh Katchekpele — 2026-01-09

Transmutation is a widely used technique for generalizing probability distributions to improve data fitting. Its implementation often relies on maximum likelihood estimation, which reduces to a box-constrained numerical optimization problem. Despite this, many studies overlook the crucial role of the initial values required to start the optimization algorithm. In this paper, we demonstrate through two case studies on real data that improper parameter initialization can lead to convergence toward local maxima, ultimately resulting in biased estimates and incorrect conclusions. We show that the choice of starting values can significantly affect both the convergence behavior and the reliability of the final results. This study highlights the need for greater methodological rigor and increased awareness regarding parameter initialization in iterative estimation procedures, particularly within the context of transmuted distributions in order to avoid erroneous conclusions.

The Role of MYB in Prostate Cancer: A Statistical Analysis

Allison Powell — 2026-01-09

Prostate cancer (PCa) continues to be a major health issue for men around the world. While treatment advances have improved outcomes for many, aggressive forms of PCa still lead to high mortality rates. Recent research highlights the MYB gene as an important contributor to tumor growth, therapy resistance, and recurrence. What is more, MYB appears to be expressed at significantly higher levels in tumors from Black men compared to White men, suggesting a possible explanation for observed racial disparities in outcomes. This paper examines MYB expression in relation to tumor progression, androgen receptor (AR) activity, and the likelihood of biochemical recurrence. Through a combination of tissue analysis, digital imaging, and public gene expression databases, we show that MYB may be a strong predictor of poor prognosis and could serve as a future target for more personalized treatment strategies.