New Nonparametric Rank-Based Tests for Paired DataReportar como inadecuado




New Nonparametric Rank-Based Tests for Paired Data - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

We propose a new nonparametric test based on therank difference between the paired sample for testing the equality of themarginal distributions from a bivariate distribution. We also consider amodification of the novel nonparametric test based on the test proposed by Baumgartern,Weiβ, and Schindler 1998. Anextensive numerical power comparison for various parametric and nonparametrictests was conducted under a wide range of bivariate distributions for smallsample sizes. The two new nonparametric tests have comparable power to thepaired t test for the data simulated from bivariate normal distributions, andare generally more powerful than the paired t test and other commonly usednonparametric tests in several important bivariate distributions.

KEYWORDS

BWS Test, Nonparametric Test, Paired Data, Power Study, Rank Difference, Wilcoxon Signed Rank Test

Cite this paper

Shan, G. 2014 New Nonparametric Rank-Based Tests for Paired Data. Open Journal of Statistics, 4, 495-503. doi: 10.4236-ojs.2014.47047.





Autor: Guogen Shan

Fuente: http://www.scirp.org/



DESCARGAR PDF




Documentos relacionados