A computational approach for testing equality of coefficients of variation in k normal populations

Abstract

In this article, a testing procedure based on computational approach testis proposed for the equality of coefficients of variation in k normal populations. We compare this procedure to some of the existing tests; the likelihood ratio, modified Bennett’s, score, generalized p-value tests in terms of the estimated type I error rates and powers by using Monte Carlo simulation. Furthermore, applications of these tests are given on a real dataset.