ABSTRACT High‐dimensional datasets have become increasingly prominent in fields such as medicine, biology, and the social sciences. These datasets often contain missing values, which can significantly compromise the reliability and validity of analytical results. This review examines simulation studies in the literature that address the issue of missing data in high‐dimensional settings. It compares the effectiveness and applicability of multiple imputation (MI) methods, discussing their strengths and weaknesses. The review also explores the challenges encountered by MI techniques in high‐dimensional data and outlines directions for future research. Its primary aim is to guide researchers and practitioners in selecting appropriate MI methods to manage missing data issues in their own datasets. The reviewed simulation studies assess the performance of various MI approaches under different missing data mechanisms and data structures, highlighting their potential advantages and limitations in real‐world applications. In conclusion, MI methods for high‐dimensional data represent a dynamic and rapidly evolving area of research. While existing approaches offer substantial capabilities, there remains a pressing need for continual innovation and rigorous inquiry to develop MI solutions that are theoretically sound, practically feasible, and capable of producing more accurate, reliable, and efficient results. This article is categorized under: Statistical Models > Simulation Models Statistical and Graphical Methods of Data Analysis > Analysis of High Dimensional Data Statistical Models > Model Selection

ABSTRACT In regression analysis, associations between continuous predictors and the outcome are often assumed to be linear. However, modeling the associations as non‐linear can improve model fit. Many flexible modeling techniques, like (fractional) polynomials and spline‐based approaches, are available. Such methods can be systematically compared in simulation studies, which require suitable performance measures to evaluate the accuracy of the estimated curves against the true data‐generating functions. Although various measures have been proposed in the literature, no systematic overview exists so far. To fill this gap, we introduce a categorization of performance measures for evaluating estimated non‐linear associations between an outcome and continuous predictors. This categorization includes many commonly used measures. The measures can not only be used in simulation studies, but also in application studies to compare different estimates to each other. We further illustrate and compare the behavior of different performance measures through some examples and a Shiny app. This article is categorized under: Statistical and Graphical Methods of Data Analysis > Modeling Methods and Algorithms

ABSTRACTInformation borrowing has been a general tool for improving estimation, testing, and prediction while learning from data distributed over several siloes (devices, trial groups, architecture, or sites), with or without preserving the privacy of the data, depending on the objectives. Federated learning is a more recent tool in this regard, which focuses on privacy preservation: this follows the concept of sharing the model parameters for a learning task rather than sharing the data, running algorithm constituents locally, and then globally optimizing parameters. There are other versions of privacy preserving information borrowing, such as personalized federated learning; the interest in such tasks can be evinced by the recent formation of a UN group to study computation measures enabling information borrowing under privacy guarantees. Heterogeneity is a bottleneck that significantly reduces the utility of information borrowing. In this paper, we identify the progress and gaps in current research regarding heterogeneity in information borrowing for federated learning. In addition to that, we also show using carefully constructed examples the limits of the federated learning paradigm when heterogeneity may be “washed out” when enough data are present. We show the limitations of current research in statistical learning and inference when heterogeneity is present.This article is categorized under: Statistical Learning and Exploratory Methods of the Data Sciences > Modeling Methods Statistical Learning and Exploratory Methods of the Data Sciences > Deep Learning Statistical and Graphical Methods of Data Analysis > Modeling Methods and Algorithms

ABSTRACTVisual statistical inference determines the significance of patterns found in data exploration through graphics. It involves human observers inspecting a lineup of plots, with one real data plot randomly placed among decoys. Each observer's cognitive skills and judiciousness can influence results. The effectiveness of this method, measured by power, depends on combining evaluations from multiple observers. Human factors influencing power, as computed by the number of detections or identifications of an observed data plot in a lineup, include observer demographics, individual skills, and experience. This paper examines these factors through studies using Amazon's Mechanical Turk, finding individual skills vary but demographics have little impact. Learning increases speed but not accuracy.This article is categorized under: Statistical Learning and Exploratory Methods of the Data Sciences > Exploratory Data Analysis Statistical and Graphical Methods of Data Analysis > Statistical Graphics and Visualization Statistical and Graphical Methods of Data Analysis > Nonparametric Methods

ABSTRACTMultivariate response linear regression model considers how a set of covariates affects multiple responses. In contrast to separately running univariate regression for each response, multivariate response regression can better estimate the coefficient matrix by exploiting shared information among the responses. A key to effectively borrowing strength across different responses is to estimate and utilize the structure among random errors in responses. Moreover, certain applications seek to understand the interrelationships of the responses, which are directly encoded in the error structure. These observations call for the development of methods for estimating the error structure in multivariate response linear regression models. This article aims at providing a review of recent progress in this challenging and important problem. We also provide simulation studies demonstrating the empirical performance of recently developed methods for error structure estimation, as well as the benefit by exploiting these estimates for regression coefficients estimation.

ABSTRACTAnalysis of longitudinal data in medical research exhibits complexities inherent to the nature of correlated and irregular‐timed observations. Functional data analysis (FDA) offers a robust framework to address these challenges, encompassing elements such as smoothing techniques, clustering adjustments, data reduction, and functional linear modeling. This systematic review was conducted under the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) statement. A comprehensive search strategy in 11 databases was conducted to identify original research manuscripts focused on FDA applications in medicine, published between January 1995 and March 2024. Manuscripts focused on such clinical applications of FDA were excluded. From 1796 potential manuscripts, 93 were selected for qualitative synthesis. Utilization of FDA methods has progressively increased over the last 25 years. Smoothing techniques (76.3%), functional principal component analysis (50.5%), supervised learning models (46.2%), and clustering methods (22.5%) were identified as the most used FDA methods. FDA represents a valuable and robust tool for analyzing longitudinal data in clinical research. Adequate application of FDA methods should improve the understanding and interpretation of subtle variations in time‐trajectories of longitudinal data sets.

ABSTRACTThe integration of statistical methods in sports sciences has become essential for decision‐making in performance analysis, injury prevention, and athlete outcomes. This work presents a systematic review following PRISMA guidelines to explore the application of ordinal regression models in the sports sciences field. A comprehensive search of articles published prior to March 4, 2023 identified 34 included studies. This search included widely recognized databases such as Web of Science, PubMed, and specialized journals of sports statistics, such as the Journal of Quantitative Analysis in Sports, and the Journal of Sports Analytics. The analysis reveals that 26.5% of these articles were published in statistics and sports statistics journals. Notably, R emerged as the primary software used for analysis in 38.8% of the studies. However, a significant majority (82.4%) of the studies did not provide data and code repositories. Among sports, soccer had the highest representation (28.6%), followed by basketball (17.1%). The most reported ordinal model was the proportional odds model (32.3%), followed by the mixed effects proportional odds model (11.8%), while a relevant proportion (29.4%) did not report the model used. Furthermore, 23.5% of articles proposed novel models. Validation tests for proportional odds models were not conducted in 53.3% of cases. This review underscores the importance of improved reporting practices, inclusivity in sport representation, and statistical education in advancing sports analytics.

image

Abstract With the development of technology and its integration with scientific realities, computer systems continue to evolve as infrastructure. One of the most important obstacles in front of quantum computers with high‐speed processing is that its existing systems cause security vulnerabilities. Therefore, in order to take advantage of quantum systems, existing systems that are already secure must also be secure in the post‐quantum scenario. One of these systems is edge computing. There are challenges in terms of computational power for the implementation of pre‐ and post‐quantum methods in structures with resource‐constrained devices. This article reviews the post‐quantum security threats of edge devices and systems and the secure methods developed for them. Although there is relatively little research in this field, it remains relevant. In the studies reviewed, lattice‐based approaches are often highlighted for making edge systems quantum‐resistant. Additionally, these studies indicate that there has been an increasing trend in this field in recent years.This article is categorized under: Applications of Computational Statistics > Defense and National Security Algorithms and Computational Methods > Networks and Security