Article
Mathematical & Computational Biology
Max Westphal, Antonia Zapf, Werner Brannath
Summary: The study presents a multiple testing framework for disease diagnostic accuracy studies with sensitivity and specificity as co-primary endpoints. It challenges the common recommendation of strict separation between model selection and evaluation, and demonstrates that evaluating multiple promising diagnostic models simultaneously can lead to better final models.
STATISTICS IN MEDICINE
(2022)
Article
Multidisciplinary Sciences
Ludivine Obry, Cyril Dalmasso
Summary: In this study, we evaluated recent weighted multiple testing procedures for genome wide association studies (GWAS) through a simulation study. We also introduced a new efficient procedure called wBHa, which prioritizes the detection of genetic variants with low minor allel frequencies while maximizing overall detection power. Our results demonstrated that wBHa outperformed other procedures in detecting rare variants while maintaining good overall power.
Article
Pharmacology & Pharmacy
John Lawrence
Summary: This study describes multiple comparisons procedures for two-armed studies with a primary hypothesis and one or more ordered secondary hypotheses. The objective is to test for an effect on the overall population and/or nonoverlapping subgroups that partition the population. The procedures ensure control of the family-wise error rate at a specified level alpha.
JOURNAL OF BIOPHARMACEUTICAL STATISTICS
(2023)
Article
Biology
Jiyang Wen, Chen Hu, Mei-Cheng Wang
Summary: This paper discusses the handling of competing risks data, particularly in clinical trials where multiple competing events may have different clinical interpretations. Estimation procedures and inferential properties are developed for the joint use of multiple cumulative incidence functions, and weighted CIFs and related metrics are considered by incorporating longitudinal marker information.
Article
Statistics & Probability
Saharon Rosset, Ruth Heller, Amichai Painsky, Ehud Aharoni
Summary: Multiple testing problems are important in modern statistical analysis, aiming to reject as many false null hypotheses as possible while controlling an overall measure of false discovery. This paper extends the optimal test for a single hypothesis to multiple testing problems and provides maximin rules for complex alternatives. The usefulness of these methods is demonstrated in numerical experiments and clinical trials, showing significant improvements in power.
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
(2022)
Article
Statistics & Probability
R. N. Montgomery, L. T. Ptomey, J. D. Mahnken
Summary: This paper builds on the recently proposed prediction test for multiple endpoints and addresses its limitations by introducing improvements and extensions. The enhanced test achieves higher power and wider applicability. An example from a physical activity trial is used to demonstrate the application of these improvements.
JOURNAL OF APPLIED STATISTICS
(2023)
Article
Business, Finance
Giulia Genoni, Piero Quatto, Gianmarco Vacca
Summary: This paper addresses the real-time detection of financial bubbles using LORD online procedures and appropriate p-value calibration, which offers new potential tools for economists to monitor and respond to bubbles in real time. The findings have significant implications for financial stability and crisis prevention. LORD algorithms are tested on globally recognized stock indexes and their effectiveness is confirmed through comparison with standard offline approaches.
FINANCE RESEARCH LETTERS
(2023)
Review
Health Care Sciences & Services
Louis Gaucher, Pierre Sabatier, Sandrine Katsahian, Anne-Sophie Jannot
Summary: This study aimed to systematically review the statistical methods used in pharmacovigilance studies without a priori hypotheses. The results showed limited use of statistical methods, with a lack of correction for multiple testing. Guidelines are recommended to improve the feasibility and accuracy of such studies.
JOURNAL OF CLINICAL EPIDEMIOLOGY
(2023)
Article
Immunology
Huajun Wang, Ellen Ypma, Uwe Nicolay
Summary: This study designed a sequential testing strategy with visualization using a graphical gatekeeping procedure to evaluate the immunogenicity of a vaccine more efficiently. The method avoids an all-or-nothing approach and highlights the importance of clinical input and agreement in designing an efficient clinical trial strategy.
Article
Computer Science, Artificial Intelligence
Vo Nguyen Le Duy, Takuto Sakuma, Taiju Ishiyama, Hiroki Toda, Kazuya Arai, Masayuki Karasuyama, Yuta Okubo, Masayuki Sunaga, Hiroyuki Hanada, Yasuo Tabei, Ichiro Takeuchi
Summary: This study proposes a novel statistical approach, called Stat-DSM, to evaluate the statistical significance of discriminative sub-trajectory mining results. The proposed method properly controls the statistical significance of the extracted sub-trajectories and addresses the computational and statistical challenges of massive trajectory datasets.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Pharmacology & Pharmacy
Yusuke Yamaguchi, Toshifumi Sugitani, Satoshi Yoshida, Kazushi Maruo
Summary: In dose-finding trials, we propose a method that combines multiple comparison procedures-modeling and individual dose-placebo comparisons in order to accelerate the drug development process while ensuring statistical power. The closed MCP-Mod and the serial gatekeeping procedures have similar statistical power and both methods aim to pursue efficacy results rather than just establishing dose-response signals.
PHARMACEUTICAL STATISTICS
(2022)
Article
Automation & Control Systems
Tijana Zrnic, Aaditya Ramdas, Michael Jordan
Summary: This study focuses on controlling the false discovery rate in asynchronous online testing, proposing a general framework that addresses dependency issues and improves existing algorithms. The use of conflict sets is highlighted as a way to better manage dependencies among test statistics.
JOURNAL OF MACHINE LEARNING RESEARCH
(2021)
Article
Statistics & Probability
James Y. Dai, Janet L. Stanford, Michael LeBlanc
Summary: Mediation analysis is gaining attention in epidemiologic studies and clinical trials, but existing methods have limitations. In this article, a multiple-testing procedure is proposed to accurately control error rates and improve power. Two data analysis examples demonstrate the effectiveness of the proposed method, and an R package is provided for implementation.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2022)
Article
Engineering, Electrical & Electronic
Prabhu Babu, Petre Stoica
Summary: In this study, the problem of smoothed nonparametric spectral estimation via cepstrum thresholding is revisited. The cepstrum thresholding problem is formulated as a multiple hypothesis testing problem, and the false discovery rate (FDR) and familywise error rate (FER) procedures are used to threshold the cepstral coefficients. The FDR and FER approaches are compared with a previously proposed individual hypothesis testing approach, and it is shown that cepstrum thresholding based on FDR and FER can yield spectral estimates with lower mean square error (MSE).
IEEE SIGNAL PROCESSING LETTERS
(2022)
Article
Mathematical & Computational Biology
Xiaodong Luo, Hui Quan
Summary: This article proposes new multiplicity adjustment procedures for clinical trials with multiple endpoints and interim analyses. The sequential procedures adapt popular multiple comparison procedures for fixed time-point design and use a-spending for each endpoint. These procedures control the family-wise Type-1 error rate and provide versatile solutions for monitoring multiple endpoints through interim analyses.
STATISTICS IN BIOPHARMACEUTICAL RESEARCH
(2023)