Abstract
In this paper, the null hypothesis is a Student's t distribution is tested. A goodness of fit (gof) test statistics involving Kullback-Leibler information (KLI) which is found based on kernel density estimation is used. The performance of the test under ranked set sampling (RSS) agianst simple random sampling (SRS) is investigated. Several alternative distributions are considered under the alternative hypothesis. Based on a simulation, it is found that the test is more efficient under RSS than SRS for the distributions considered.
