Rank-Based Procedures in Factorial Designs: Hypotheses About Non-Parametric Treatment Effects

Edgar Brunner1, Frank Konietschke2, Markus Pauly3, Madan L. Puri4
1University of Göttingen, Germany
2University of Texas, Dallas, USA
3University of Ulm, Germany
4Indiana University Bloomington, USA

Tóm tắt

Summary Existing tests for factorial designs in the non-parametric case are based on hypotheses formulated in terms of distribution functions. Typical null hypotheses, however, are formulated in terms of some parameters or effect measures, particularly in heteroscedastic settings. Here this idea is extended to non-parametric models by introducing a novel non-parametric analysis-of-variance type of statistic based on ranks or pseudoranks which is suitable for testing hypotheses formulated in meaningful non-parametric treatment effects in general factorial designs. This is achieved by a careful detailed study of the common distribution of rank-based estimators for the treatment effects. Since the statistic is asymptotically not a pivotal quantity we propose three different approximation techniques, discuss their theoretic properties and compare them in extensive simulations together with two additional Wald-type tests. An extension of the presented idea to general repeated measures designs is briefly outlined. The rank- and pseudorank-based procedures proposed maintain the preassigned type I error rate quite accurately, also in unbalanced and heteroscedastic models.

Từ khóa


Tài liệu tham khảo

Acion, 2006, Probabilistic index: an intuitive non-parametric approach to measuring the size of treatment effects, Statist. Med., 25, 591, 10.1002/sim.2256

Akritas, 1994, Fully non-parametric hypotheses for factorial designs I: Multivariate repeated measures designs, J. Am. Statist. Ass., 89, 336, 10.1080/01621459.1994.10476475

Akritas, 1997, Nonparametric hypotheses and rank statistics for unbalanced factorial designs, J. Am. Statist. Ass., 92, 258, 10.1080/01621459.1997.10473623

Akritas, 1997, A unified approach to ranks tests in mixed models, J. Statist. Planng Inf., 61, 249, 10.1016/S0378-3758(96)00177-2

Bamber, 1975, The area above the ordinal dominance graph and the area below the receiver operating characteristic graph, J. Math. Psychol., 12, 387, 10.1016/0022-2496(75)90001-2

Boos, 1992, A rank-based mixed model approach to multisite clinical trials, Biometrics, 44, 61, 10.2307/2532739

Box, 1954, Some theorems on quadratic forms applied in the study of analysis of variance problems, I: Effect of inequality of variance in the one-way classification, Ann. Math. Statist., 25, 290, 10.1214/aoms/1177728786

Brown, 2002, Kruskal-Wallis, multiple comparisons and Efron dice, Aust. New Zeal. J. Statist., 44, 427, 10.1111/1467-842X.00244

Brumback, 2006, Using the ROC curve for gauging treatment effect in clinical trials, Statist. Med., 25, 575, 10.1002/sim.2345

Brunner, 1997, Box-type approximations in non-parametric factorial designs, J. Am. Statist. Ass., 92, 1494, 10.1080/01621459.1997.10473671

Brunner, 2000, Nonparametric analysis of ordered categorical data in designs with longitudinal observations and small sample sizes, Biometr. J., 42, 663, 10.1002/1521-4036(200010)42:6<663::AID-BIMJ663>3.0.CO;2-7

Brunner, 2000, The nonparametric Behrens-Fisher problem: asymptotic theory and a small-sample approximation, Biometr. J., 42, 17, 10.1002/(SICI)1521-4036(200001)42:1<17::AID-BIMJ17>3.0.CO;2-U

Brunner, 2013, Nichtparametrische Datenanalyse, 10.1007/978-3-642-37184-4

Brunner, 1999, Rank-score tests in factorial designs with repeated measures, J. Multiv. Anal., 70, 286, 10.1006/jmva.1999.1821

Brunner, 2001, Nonparametric methods in factorial designs, Statist. Pap., 42, 1, 10.1007/s003620000039

Brunner, 2013, Handbook of Methods and Applications of Statistics in Clinical Trials, 471

De Neve, 2014, The unified Wilcoxon-Mann-Whitney test for analyzing RT-qPCR data in R, Bioinformatics, 30, 2494, 10.1093/bioinformatics/btu313

De Neve, 2015, A regression framework for rank tests based on the Probabilistic Index Model, J. Am. Statist. Ass., 110, 1276, 10.1080/01621459.2015.1016226

DiCiccio, 2015, Robust permutation tests for correlation and regression coefficients

Domhof, 2001, Nichtparametrische relative Effekte

Fan, 2014, Wald-type rank tests: a GEE approach, Computnl Statist. Data Anal., 74, 1, 10.1016/j.csda.2013.12.004

Fan, 2015, On power and sample size of the ANOVA-type rank test, Communs Statist. Simuln Computn

Fischer, 2015, Mann-Whitney type tests for microarray experiments: the R package gMWT, J. Statist. Softwr., 65, 1

Fischer, 2014, Generalized Mann-Whitney type tests for microarray experiments, Scand. J. Statist., 41, 672, 10.1111/sjos.12055

Gao, 2005, A unified non-parametric approach for unbalanced factorial designs, J. Am. Statist. Ass., 100, 926, 10.1198/016214505000000042

Gao, 2008, Nonparametric multiple comparison procedures for unbalanced two-way layouts, J. Statist. Planng Inf., 138, 3674, 10.1016/j.jspi.2007.11.019

Gao, 2008, Nonparametric multiple comparison procedures for unbalanced one-way factorial designs, J. Statist. Planng Inf., 138, 2574, 10.1016/j.jspi.2007.10.015

Gardner, 1970, The paradox of the nontransitive dice and the elusive principle of indifference, Scient. Am., 223, 110, 10.1038/scientificamerican1270-110

Janssen, 2003, How do bootstrap and permutation tests work?, Ann. Statist., 31, 768, 10.1214/aos/1056562462

Kaufmann, 2005, Nonparametric methods for analysing the accuracy of diagnostic tests with multiple readers, Statist. Meth. Med. Res., 14, 129, 10.1191/0962280205sm392oa

Kieser, 2013, Assessment of statistical significance and clinical relevance, Statist. Med., 32, 1707, 10.1002/sim.5634

Konietschke, 2010, Testing and estimation of purely non-parametric effects in repeated measures designs, Computnl Statist. Data Anal., 54, 1895, 10.1016/j.csda.2010.02.019

Konietschke, 2012, Rank-based multiple test procedures and simultaneous confidence intervals, Electron. J. Statist., 6, 738, 10.1214/12-EJS691

Kruskal, 1952, A nonparametric test for the several sample problem, Ann. Math. Statist., 23, 525, 10.1214/aoms/1177729332

Kruskal, 1952, Use of ranks in one-criterion variance analysis, J. Am. Statist. Ass., 47, 583, 10.1080/01621459.1952.10483441

Lange, 2008, Nichtparametrische Modelle für faktorielle Diagnosestudien

Mann, 1947, On a test of whether one of two random variables is stochastically larger than the other, Ann. Math. Statist., 18, 50, 10.1214/aoms/1177730491

Mathai, 1992, Quadratic Forms in Random Variables

Munzel, 1999, Linear rank score statistics when ties are present, Statist. Probab. Lett., 41, 389, 10.1016/S0167-7152(98)00185-0

Pauly, 2015, Asymptotic permutation tests in general factorial designs, J. R. Statist. Soc., 77, 461, 10.1111/rssb.12073

Pauly, 2015, Analysis of high-dimensional one group repeated measures designs, Statistics, 49, 1243, 10.1080/02331888.2015.1050022

Puri, 1964, Asymptotic efficiency of a class of c-sample tests, Ann. Math. Statist., 35, 102, 10.1214/aoms/1177703733

R Development Core Team, 2010, R: a Language and Environment for Statistical Computing

Rust, 1984, A modification of the Kruskal-Wallis statistic for the generalized Behrens-Fisher problem, Communs Statist. Theory Meth., 13, 2007, 10.1080/03610928408828810

Ruymgaart, 1980, Statistique non Parametrique Asymptotique, 1

Thangavelu, 2007, Wilcoxon–Mann–Whitney test for stratified samples and Efron's paradox dice, J. Statist. Planng Inf., 137, 720, 10.1016/j.jspi.2006.06.005

Thas, 2012, Probabilistic index models (with discussion), J. R. Statist. Soc., 74, 623, 10.1111/j.1467-9868.2011.01020.x

Umlauft, 2016, Rank-based permutation approaches for nonparametric designs, Br. J. Math. Statist. Psychol.

Vallejo, 2010, Analysis of unbalanced factorial designs with heteroscedastic data, J. Statist. Computn Simuln, 80, 75, 10.1080/00949650802482386

Vermeulen, 2015, Increasing the power of the Mann-Whitney test in randomized experiments through flexible covariate adjustment, Statist. Med., 34, 1012, 10.1002/sim.6386

Zapf, 2015, Nonparametric meta-analysis for diagnostic accuracy studies, Statist. Med., 34, 3831, 10.1002/sim.6583