Distribution-free two-sample homogeneity test for circular data based on geodesic distance

Document Type : Research Paper

Authors
Ahmed Jebur Ali
Samira Faisal Abushilah
Department of Mathematics, College of Education for Girls, University of Kufa, Najaf, Iraq
10.22075/ijnaa.2022.5992
Abstract
A new distance-based homogeneity test for circular data is considered in this paper. This test statistic could be used with a permutation test to detect the homogeneity between the distributions of two angular groups. A comparison between the proposed test against the Randomisation Watson test (RWT) and Wheeler Watson test (WWT) is addressed as well. The power of the new test has been computed and compared against RWT and WWT powers based on the angular simulated data which have been generated from the von Mises distributions. The simulation study based on the power demonstrates that the proposed statistical test outperforms classical tests.
Keywords
Circular Data
Geodesic Distance
Von Mises Distribution
Randomisation Watson Test
Wheeler Watson Test
Energy Statistic
[1] F.H. Clarke, Y.S. Ledyaev, R.J. Stern and P.R. Wolenski, Nonsmooth Analysis and Control Theory, Springer Science and Business Media, 2008.
[2] H.Cramer, On the composition of elementary errors, Scand. Actuarial J. 1928(1) (2011) 13—74.
[3] S.R. Jammalamadaka and A. Sengupta, Topics in Circular Statistics, Volume 5, World Scientific, 2001.
[4] L. Landler, G.D. Ruxton and E.P. Malkemper, Circular data in biology: advice for effectively implementing statistical procedures, Behav. Eco.Sociobio. 72 (2018) 1—10.
[5] K. Mardia, A non-parametric test for the bivariate two-sample location problem, J. Royal Stat. Soc. Ser. B 29 (1967) 320-–342.
[6] K. Mardia, On Wheeler and Watson’s two-sample test on a circle, Indian J. Stat. Ser. A 31(2) (1969) 177-–190.
[7] K.V. Mardia and P.E. Jupp, Directional Statistics, volume 494, John Wiley & Sons, 2009.
[8] M.L. Rizzo and G.J. Szekely, Energy distance, Wiley Interdisciplinary Reviews: Comput. Stat. 8 (2016) 27—38.
[9] G.J. Szekely and M.L. Rizzo, Energy statistics: A class of statistics based on distances, J. Stat. Plan. Infer. 143 (2013) 1249—1272.
[10] S. Wheeler and G.S. Watson, A distribution-free two-sample test on a circle, Biometrika 51(1-2) (1964) 256–257.
[11] J.H. Zar, Biostatistical Analysis, Pearson, 2014.
International Journal of Nonlinear Analysis and Applications
Volume 13, Issue 1
March 2022
Pages 2703-2711

  • Receive Date 08 September 2021
  • Revise Date 18 October 2021
  • Accept Date 11 December 2021

Article View 19,984
PDF Download 10,412