Bootstrapping and permuting paired t-test type statistics

Konietschke, Frank; Pauly, Markus

doi:10.1007/s11222-012-9370-4

Cited by 71 publications

(48 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further, resampling data with replacement from the pooled data would increase more resampling variability than resampling data with replacement within each group separately under matched/paired design. A recent study referred this pooled approach as a counterintuitive resampling strategy for paired group comparison and showed asymptotically valid for comparing means [23]. In this study, the null hypothesis being tested is that the means of underlying distributions are the same, that is, H 0 : μ 1 = μ 2 for comparison of two groups or H 0 : μ1 = μ2 = μ3 for comparison of three groups, where μ i is the mean for i th group, i = (1,2,3).…”

mentioning

confidence: 99%

Analysis of small sample size studies using nonparametric bootstrap test with pooled resampling method

Dwivedi

Mallawaarachchi

Alvarado

2017

Statistics in Medicine

226

138

View full text Add to dashboard Cite

Experimental studies in biomedical research frequently pose analytical problems related to small sample size. In such studies, there are conflicting findings regarding the choice of parametric and nonparametric analysis, especially with non-normal data. In such instances, some methodologists questioned the validity of parametric tests and suggested nonparametric tests. In contrast, other methodologists found nonparametric tests to be too conservative and less powerful and thus preferred using parametric tests. Some researchers have recommended using a bootstrap test; however, this method also has small sample size limitation. We used a pooled method in nonparametric bootstrap test that may overcome the problem related with small samples in hypothesis testing. The present study compared nonparametric bootstrap test with pooled resampling method corresponding to parametric, nonparametric, and permutation tests through extensive simulations under various conditions and using real data examples. The nonparametric pooled bootstrap t-test provided equal or greater power for comparing two means as compared with unpaired t-test, Welch t-test, Wilcoxon rank sum test, and permutation test while maintaining type I error probability for any conditions except for Cauchy and extreme variable lognormal distributions. In such cases, we suggest using an exact Wilcoxon rank sum test. Nonparametric bootstrap paired t-test also provided better performance than other alternatives. Nonparametric bootstrap test provided benefit over exact Kruskal-Wallis test. We suggest using nonparametric bootstrap test with pooled resampling method for comparing paired or unpaired means and for validating the one way analysis of variance test results for non-normal data in small sample size studies. Copyright © 2017 John Wiley & Sons, Ltd.

show abstract

mentioning

confidence: 99%

Analysis of small sample size studies using nonparametric bootstrap test with pooled resampling method

Dwivedi

Mallawaarachchi

Alvarado

2017

Statistics in Medicine

226

138

View full text Add to dashboard Cite

show abstract

“…However, the number of assessed embryos should not be neglected. We thus used a random permutation testing approach[37], accounting for these challenges and opportunities, to assess the significance of differences in embryopathy rates between matched left and right uterine horn pairs, across mothers. This approach has the advantages of (i) stemming naturally from the natural pairing between uterine horns, (ii) not requiring a specific sample distribution, and (iii) accounting for the inherently unbalanced pairing between uterine horns.…”

Section: Methodsmentioning

confidence: 99%

Hyperglycemia induces embryopathy, even in the absence of systemic maternal diabetes: An in vivo test of the fuel mediated teratogenesis hypothesis

Baack

Wang

et al. 2014

Reproductive Toxicology

View full text Add to dashboard Cite

Embryonic exposure to excess circulating fuels is proposed to underlie diabetic embryopathy. To isolate the effects of hyperglycemia from the many systemic anomalies of diabetes, we infused 4 mg/min glucose into the left uterine artery of non-diabetic pregnant rats on gestation days (GD) 7–9. Right-sided embryos and dams exhibited no glucose elevation. Embryos were assessed on GD13, comparing the left versus right uterine horns. Hyperglycemic exposure increased rates of embryopathy, resorptions, and worsened embryopathy severity. By contrast, saline infusion did not affect any of these parameters. To assess for possible embryopathy susceptibility bias between uterine horns, separate dams were given retinoic acid (25 mg/kg, a mildly embryopathic dose) systemically on GD7.5. The resultant embryopathy rates were equivalent between uterine horns. We conclude that hyperglycemia, even in the absence of systemic maternal diabetes, is sufficient to produce in vivo embryopathy during organogenesis.

show abstract

“…Janssen (, ) proposes studentized permutation tests for the parametric Behrens–Fisher problem, whereas Fay and Proschan () discuss test recommendations in both Behrens–Fisher problems. Moreover, Janssen (,b), Pauly (), Konietschke and Pauly (, ), Omelka and Pauly (), Chung and Romano (, ), as well as Pauly et al. () support the use of studentized permutation tests for other testing problems.…”

Section: Introductionmentioning

confidence: 93%

Permutation‐based inference for the AUC: A unified approach for continuous and discontinuous data

2016

Self Cite

View full text Add to dashboard Cite

We investigate rank-based studentized permutation methods for the nonparametric Behrens-Fisher problem, that is, inference methods for the area under the ROC curve. We hereby prove that the studentized permutation distribution of the Brunner-Munzel rank statistic is asymptotically standard normal, even under the alternative. Thus, incidentally providing the hitherto missing theoretical foundation for the Neubert and Brunner studentized permutation test. In particular, we do not only show its consistency, but also that confidence intervals for the underlying treatment effects can be computed by inverting this permutation test. In addition, we derive permutation-based range-preserving confidence intervals. Extensive simulation studies show that the permutation-based confidence intervals appear to maintain the preassigned coverage probability quite accurately (even for rather small sample sizes). For a convenient application of the proposed methods, a freely available software package for the statistical software R has been developed. A real data example illustrates the application.

show abstract

Bootstrapping and permuting paired t-test type statistics

Cited by 71 publications

References 26 publications

Analysis of small sample size studies using nonparametric bootstrap test with pooled resampling method

Analysis of small sample size studies using nonparametric bootstrap test with pooled resampling method

Hyperglycemia induces embryopathy, even in the absence of systemic maternal diabetes: An in vivo test of the fuel mediated teratogenesis hypothesis

Permutation‐based inference for the AUC: A unified approach for continuous and discontinuous data

Contact Info

Product

Resources

About