This paper takes a close look at balanced permutations, a recently developed sample reuse method with applications in bioinformatics. It turns out that balanced permutation reference distributions do not have the correct null behavior, which can be traced to their lack of a group structure. We find that they can give p-values that are too permissive to varying degrees. In particular the observed test statistic can be larger than that of all B balanced permutations of a data set with a probability much higher than 1/(B + 1), even under the null hypothesis.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3148117 | PMC |
| http://dx.doi.org/10.1089/cmb.2008.0144 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Predicting Hospitalization in Older Adults Using Machine Learning.
Geriatrics (Basel)
January 2025
School of Telematics, University of Colima, Colima 28040, Mexico.
: Hospitalization among older adults is a growing challenge in Mexico due to the high prevalence of chronic diseases and limited public healthcare resources. This study aims to develop a predictive model for hospitalization using longitudinal data from the Mexican Health and Aging Study (MHAS) using the random forest (RF) algorithm. : An RF-based machine learning model was designed and evaluated under different data partition strategies (ST) with and without variable interaction.
View Article and Find Full Text PDFAnalysis methods for covariate-constrained cluster randomized trials with time-to-event outcomes.
BMC Med Res Methodol
January 2025
Department of Biostatistics and Bioinformatics, Emory University, Atlanta, Georgia, USA.
Background: Cluster randomized trials, which often enroll a small number of clusters, can benefit from constrained randomization, selecting a final randomization scheme from a set of known, balanced randomizations. Previous literature has addressed the suitability of adjusting the analysis for the covariates that were balanced in the design phase when the outcome is continuous or binary. Here we extended this work to time-to-event outcomes by comparing two model-based tests and a newly derived permutation test.
View Article and Find Full Text PDFPredicting Antidepressant Treatment Response From Cortical Structure on MRI: A Mega-Analysis From the ENIGMA-MDD Working Group.
Hum Brain Mapp
January 2025
Amsterdam UMC, Department of Radiology and Nuclear Medicine, University of Amsterdam, Amsterdam, the Netherlands.
Accurately predicting individual antidepressant treatment response could expedite the lengthy trial-and-error process of finding an effective treatment for major depressive disorder (MDD). We tested and compared machine learning-based methods that predict individual-level pharmacotherapeutic treatment response using cortical morphometry from multisite longitudinal cohorts. We conducted an international analysis of pooled data from six sites of the ENIGMA-MDD consortium (n = 262 MDD patients; age = 36.
View Article and Find Full Text PDFUsing Social Network Analysis to Assess 'Groupness' in a Mixed-Species Zoo Exhibit of Tufted Capuchins () and Squirrel Monkeys ().
Animals (Basel)
November 2024
Psychology, Faculty of Natural Sciences, University of Stirling, Stirling FK9 4LA, UK.
Mixed-species groups have been recorded in various primates, including tufted capuchin and squirrel monkeys. Measures of their 'groupness' are typically based on factors such as group stability, social interactions, proximity, or behavioural coordination. Social network analysis has become a useful tool for quantifying relationships among group-living individuals.
View Article and Find Full Text PDFBetter power by design: Permuted-subblock randomization boosts power in repeated-measures experiments.
Psychol Methods
December 2024
School of Psychology & Neuroscience, University of Glasgow.
Article Synopsis
- - Participants in experiments adapt due to factors like learning and fatigue, which can influence measurement accuracy and reduce statistical power.
- - A new method, called permuted-subblock randomization (PSR), helps improve statistical power by balancing conditions throughout an experimental session, demonstrated through Monte Carlo simulations.
- - PSR increased power by an average of 13%, with some designs seeing boosts up to 45%, while maintaining control over false positives when there was no time-dependent variation; an R package named "explan" is available for implementing this method.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!