A nonparametric test for scale in univariate population two-sample setup

Article type: Research Article

Authors: Mathur, Sunil^{; *} | Dolo, Samuel

Affiliations: Department of Mathematics, University of Mississippi, University, MS 38677-1848, USA

Correspondence: [*] Corresponding author. E-mail: skmathur@olemiss.edu.

Abstract: A new method is proposed for testing the equality of two populations in a scale model. The power of the proposed test is compared using the Monte Carlo simulation technique with the Siegel-Tukey test [26], the Klotz score test [16], and Levene's test [18]. Simulation studies show that the proposed test performs better than Siegel-Tukey's test, the Klotz score test, and Levene's test under almost all the distributions considered in the study. We calculate the asymptotic relative efficiency of the proposed test. We apply the proposed test to a real life situation.

Keywords: Test, scale, rank, data, efficiency, power

DOI: 10.3233/MAS-2007-2305

Journal: Model Assisted Statistics and Applications, vol. 2, no. 3, pp. 145-152, 2007