Comparison of three permutation test algorithms applied to the multivariate means of two independent samples
DOI:
https://doi.org/10.26439/ing.ind2019.n037.4545Abstract
The objective of this research was to compare three permutation test algorithms. Data scenarios obtained through the Monte Carlo simulation were proposed and the suggested algorithms were applied to each of them. The results showed a test power greater than 0,85. The first algorithm based on the Hotelling’s T2 presented the highest test power. The implementation of the algorithms was carried out using the R statistical program.
Downloads
References
Blair, C., Higgins, J., Karniski ,W. y Kromrey, J. (2010). A Study of Multivariate Permutation Tests Which May Replace Hotelling’s T2 Test in Prescribed Circumstances. Journal Multivariate Behavioral Research 29(2), pp. 141-163. doi: 10.1207/ s15327906mbr2902_2
Butar, F. y Park, J. (2008). Permutation Test for Comparing Two Populations. Journal of Mathematical Sciences & Mathematics Education 3(2), pp. 19-30.
Chung, E. y Romano, J. (2011). Asymptotically valid and exact permutation tests based on twosample U-statistics. (Technical report No. 2011-09). Stanford: Stanford University. Recuperado de https://statistics.stanford.edu/sites/default/files/2011-09.pdf
Chung, E. y Romano, J. (2013). Multivariate and Multiple Permutation Test. (Technical report No.2013-05). Stanford: Stanford University. Recuperado de https://statistics. stanford.edu/sites/g/files/sbiybj6031/f/2013-05_0.pdf
Efron, B. y Tibshirani, R. (2011): An Introduction to the Bootstrap. Nueva York: Chapman & Hall/CRC.
Einsporn, R. y Habtzghi, D. (2013): Combining Paired and Two-Sample Data Using a Permutation. Journal of Data Science 11, pp. 767-779.
Higgins, J. (2004). An introduction to modern nonparametric statistics. Londres: Thomson Learning.
Samuh, M. (2017). Ranked Set Two Sample Permutation Test. Statistica 3, pp. 237-249. The R Project for Statistical Computing (3.6) [Software]. (2019). Recuperado de https://
www.r-project.org/
