Nonparametric hypothesis testing for equality of means on the simplex

Tsagris, Michail; Preston, Simon P.; Wood, Andrew T.A.

Abstract

In the context of data that lie on the simplex, we investigate use of empirical and exponential empirical likelihood, and Hotelling and James statistics, to test the null hypothesis of equal population means based on two independent samples. We perform an extensive numerical study using data simulated from various distributions on the simplex. The results, taken together with practical considerations regarding implementation, support the use of bootstrap-calibrated James statistic.

Citation

Tsagris, M., Preston, S. P., & Wood, A. T. (in press). Nonparametric hypothesis testing for equality of means on the simplex. Journal of Statistical Computation and Simulation, 87(2), https://doi.org/10.1080/00949655.2016.1216554

• Journal Article Type: Article
• Acceptance Date: Jul 19, 2016
• Online Publication Date: Aug 2, 2016
• Deposit Date: Jun 30, 2017
• Publicly Available Date: Jun 30, 2017
• Journal: Journal of Statistical Computation and Simulation
• Print ISSN: 0094-9655
• Electronic ISSN: 1563-5163
• Publisher: Taylor and Francis
• Peer Reviewed: Peer Reviewed
• Volume: 87
• Issue: 2
• DOI: https://doi.org/10.1080/00949655.2016.1216554
• Keywords: Compositional data, hypothesis testing, Hotelling test, James test, nonparametric, empirical likelihood, bootstrap
• Public URL: https://nottingham-repository.worktribe.com/output/806922
• Publisher URL: http://www.tandfonline.com/doi/abs/10.1080/00949655.2016.1216554
• Additional Information: This is an Accepted Manuscript of an article published by Taylor & Francis in Journal of Statistical Computation and Simulation on 02 August 2016, available online: http://www.tandfonline.com/10.1080/00949655.2016.1216554
• Contract Date: Jun 30, 2017