Abstract
Analysis of variance (ANOVA) is used to compare the means of various samples. Parametric ANOVA approaches assume normally distributed error terms within subsamples. Permutation tests like synchronized permutation tests are computationally intensive and distribution free procedures. Hence they overcome the limitations of parametric methods. Unbalanced designs with differing subsample sizes are quite frequent in various disciplines. There is a broad literature about unbalanced designs and parametric testing. For permutation tests this topic received some attention recently. This paper extends the synchronized permutation method to unbalanced two-level ANOVA designs. A simulation study investigates the behavior of different procedures for various types of unbalanced designs. It compares the results to other permutation approaches. The synchronized permutation method yields comparable results to the best performing competing permutation approaches. However the approach is limited to certain kinds of unbalanced designs.
Similar content being viewed by others
References
Akritas MG, Arnold SF, Brunner E (1997) Nonparametric hypotheses and rank statistics for unbalanced factorial designs. J Am Stat Assoc 92(437):258–265. doi:10.2307/2291470
Anderson MJ, ter Braak CJF (2003) Permutation tests for multi-factorial analysis of variance. J Stat Comput Simul 73(2):85–113. doi:10.1080/0094965021000015558
Basso D, Chiarandini M, Salmaso L (2007) Synchronized permutation tests in replicated I \(x\) J designs. J Stat Plan Inference 137:2564–2578. doi:10.1016/j.jspi.2006.04.016
Basso D, Pesarin F, Salmaso L, Solari A (2009) Permutation tests for stochastic ordering and ANOVA. Springer, New York
Brunner E, Puri ML (2001) Nonparametric methods in factorial designs. Stat Pap 42:1–52
Corain L, Salmaso L (2007) A critical review and a comparative study on conditional permutation tests for two-way ANOVA. Commun Stat-Simul Comput 36(4):791–805. doi:10.1080/03610910701418119
Edgington ES (1995) Randomization tests. Statistics textbooks and monographs, 3rd edn. Marcel Dekker, New York
Everitt BS, Hothorn T (2009) A handbook of statistical analyses using R, 2nd edn. Chapman and Hall, Boca Raton
Good P (2005) Permutation, parametric, and bootstrap tests of hypotheses. Springer series in statistics, 3rd edn. Springer, New York
Hand DJ, Daly F, Lunn AD, McConway KJ, Ostrowski E (1994) A handbook of small datasets. Chapman & Hall, London
Kherad-Pajouh S, Renaud D (2010) An exact permutation method for testing any effect in balanced and unbalanced fixed effect ANOVA. Comput Stat Data Anal 54:1881–1893. doi:10.1016/j.csda.2010.02.015
Manly BFJ (1997) Randomization, bootstrap and Monte Carlo methods in biology. Texts in statistical science, 2nd edn. Chapman & Hall, London
Manly BFJ (2007) Randomization, bootstrap and Monte Carlo methods in biology. Texts in statistical science series, 3rd edn. CRC Press INC, New York
Mendes M, Akkartal E (2010) Comparison of ANOVA F and WELCH tests with their respective permutation versions in terms of Type I error rates and test power. Kafas Univ Vet Fak Derg 16(5):711–716
Pauly M, Brunner E, Konietschke F (2014) Asymptotic permutation tests in general factorial designs. J R Stat Soc: Ser B 77(2):461–473. doi:10.1111/rssb.12073
Pesarin F (2001) Multivariate permutation tests with applications in biostatistics. Wiley & Sons, Chichester
Pesarin F, Salmaso L (2010) Permutation tests for complex data: theory. Application and software. John Wiley & Sons Ltd, New York
R Core Team (2013) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna
Salmaso L (2003) Synchronized permutation tests in \(2^{k}\) factorial designs. Commun Stat 32:1419–1437. doi:10.1081/STA-120021566
Searle SR (1987) Linear models for unbalanced data. Wiley series in probability and mathematical statistics. John Wiley & Sons Ltd, New York
Acknowledgments
The authors gratefully acknowledge the syntax for the WTPS approach provided by Frank Konietschke.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1: Comparing restricted and simple weights
This appendix explains why the restricted weights approach and the fixed weights approach do lead to the same results in the simple unbalanced and the crossed unbalanced case.
The weights in the restricted weights approach are:
Note that we can compute these weights only when the denominator is not zero.
As in the simple unbalanced design \(n_{11}=n_{12}\) and \(n_{21}=n_{21}\), Eq. 26 can be written in the following way:
As all cell frequencies are positive and if the denominator is not zero, Eq. 31 simplifies to
We can apply the same procedure to the other weights and also to the weights when imposing the sample size restrictions of the crossed unbalanced design. For both cases we can finally write the weights in the following way:
When we compare Eqs. 32–35 to the simple weights approach (\(w_{11}^{*}=n_{21}\), \(w_{12}^{*}=n_{22}\), \(w_{21}^{*}=n_{11}\), and \(w_{22}^{*}=n_{12}\)), it is obvious that they are identical except for the denominator that is not present in the simple weights approach. As the denominator is the total sample size, it is constant for every permutation. Hence, the distribution of the permuted test statistic differs only because of this constant. The order of the permuted test statistics does not change, and so the achieved p value the same. A further consideration from this is that the simple weights approach should be used in the simple unbalanced and the crossed unbalanced design as it yields the same results when the denominator of the restricted weights approach is not zero and it is also applicable in cases when the denominator is zero.
Appendix 2: Extended simulation results
During the review process we were asked to include some further simulation results on more data settings. Due to saving space in the article, these are provided in this appendix.
We extended the second part of the previous simulation study with regard to the following aspects:
-
More subsample size settings: This was realized by including subsample sizes that differed by no, two, five, and nine observations
-
Effect sizes: We included settings in which the respective effect was inactive. In others the effect was slightly lower and on others it was slightly larger than in the previous simulations. We selected the effect sizes in a way that the procedures did not achieve a too high power because in this case, differences the behavior of the procedures is almost indistinguishable.
Further, we included the interaction effect as an effect of interest.
We present the results for testing the main effect and the interaction term separately. Each plot contains the results for different minimal subsample sizes, different active effects and different error term distributions. Further, the plots contain the results when there was no effect (solid lines) and when there was an active effect (dashed lines for the smaller effect and dotted lines for the larger effect) in the data.
For the main effect A, Fig. 3 presents the results for equally sized samples, Fig. 4 presents the results for samples where the difference between the subsample sizes was two observations, and Figs. 5 and 6 correspond to settings in which the difference in subsample sizes was five and nine observations, respectively.
Similarly, for the interaction effect, Fig. 7 presents the results for equally sized samples, and Figs. 8, 9, and 10 correspond to settings in which the difference in subsample sizes were two, five and nine observations, respectively.
The pattern throughout the different settings is similar to the findings in the article: Both, the CSP and the USP algorithm need sufficiently large sample sizes to work properly. From the new results it becomes visible that a larger total sample size does improve the power behavior of the USP approach. The explanation for this is that the number of possible permutation increases.
Rights and permissions
About this article
Cite this article
Hahn, S., Salmaso, L. A comparison of different synchronized permutation approaches to testing effects in two-level two-factor unbalanced ANOVA designs. Stat Papers 58, 123–146 (2017). https://doi.org/10.1007/s00362-015-0690-2
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00362-015-0690-2