The Likelihood Ratio Test Based On Ranked Set Sampling

Ranked set sampling is an approach to more efficient data collection, People pay more and more attention to the statistic inference using this sampling. There are many results in parametric estimators and nonparametric inference, while few scholars discuss the tests of parametric hypotheses. In this paper, we consider the Likelihood ratio test statistic using ranked set sampling from the simple and composite hypothesis. It is proved that the test statistic has the same asymptotic distribution as the test statistic using simple set sampling under the null hypothesis and regularity condition. Next, we compared with the power of the test based on RSS and SRS in the subset of parameter and explain the likelihood ratio test based on RSS is more efficient from the faction of two errors .