An Optimal Nonparametric Test of Symmetry Based on the Overlapping Coefficient Using an Extreme Ranked Set Sample: Application to Noninvasive Measurement of Cardiac Output by Electrical Velocimetry in Neonates

Hani M. Samawi, Georgia Southern University
Robert L. Vogel, Georgia Southern University
Barbara Weaver
Joseph M. Van De Water, Mercer University


In this paper we provide a more efficient nonparametric test of symmetry based on the empirical overlap coefficient using kernel density estimation applied to an extreme ranked set sample. Our investigation reveals that our proposed test of symmetry is more powerful than all available tests of symmetry. Intensive simulation is conducted to examine the power of the proposed test. An illustration is provided using cardiac output and body weight of neonates in a neonatal intensive care unit.