Comparison of Tests for the Equality of Several Log-Normal Means

Abstract

In this paper, we focused on testing for the equality of several log-normal means since the log-normal distribution is one of the most common distribution for analyzing positive and right-skewed data. Recently, many researchers proposed a lot of methods based on likelihood-based methods, generalized pivotal-based methods, bootstrap-based methods for this case. Apparently because there is not exact result related with which test is better than others in which cases, our goal shed light on this important problem. For this reason, we investigate these methods and compare them with each other by simulation study.