Saeid Amiri: On the Application of the Transformation to Testing the Coefficient of Variation

It is difficult to drive mathematically the theoretical sampling distribution of the coefficient of variation (CV) and to make inferences about it. This presentation attempts to provide an overview of a parametric asymptotic inference of the coefficient of variation using a transformation that gives variance stabilization. Although it can easily be shown that the variance of the logarithm of the sample mean is approximately the coefficient of variation, up to now it has not been demonstrated how this idea can be used to draw inferences concerning the CV. This article shows how the bootstrap method can be used to improve the discussed methods and deals with one- and two-sample tests of the CV.