Small sample sizes: A big data problem in high-dimensional data analysis. (March 2021)

• Record Type: Journal Article
• Authors: Konietschke, Frank; Schwab, Karima; Pauly, Markus
• Is Part Of: Statistical methods in medical research. Volume 30:Number 3(2021)
• Page Start: 687
• Page End: 701
• Publication Date: 2021-03
• Subjects: Multiple contrast tests -- max t-test -- repeated measures -- resampling -- simultaneous confidence intervals; Medicine -- Research -- Statistical methods -- Periodicals; Research -- Periodicals; Review Literature -- Periodicals; Statistics -- methods -- Periodicals; Médecine -- Recherche -- Méthodes statistiques -- Périodiques; 610.727
• DOI: 10.1177/0962280220970228
• Languages: English
• ISSNs: 0962-2802

Abstract

In many experiments and especially in translational and preclinical research, sample sizes are (very) small. In addition, data designs are often high dimensional, i.e. more dependent than independent replications of the trial are observed. The present paper discusses the applicability of max t -test-type statistics (multiple contrast tests) in high-dimensional designs (repeated measures or multivariate) with small sample sizes. A randomization-based approach is developed to approximate the distribution of the maximum statistic. Extensive simulation studies confirm that the new method is particularly suitable for analyzing data sets with small sample sizes. A real data set illustrates the application of the methods.