New Tests to Detect Outliers in the Pareto Distribution T. Hosseini 1 and M. Jabbari Nooghabi 2 Department of Statistics, Ferdowsi University of Mashhad, Mashhad-Iran Abstract In this article, we investigate outlier observations in the Pareto distribution, by introducing two new tests: the generalized likelihood ratio and the uniformly most powerful tests for the outlier parameter of this distribution. Prior to this, we out- line the necessary prerequisites for our study, including a model for outliers, the density function of the Pareto distribution in the presence of k outliers which is obtained from the same distribution, etc. Furthermore, we present joint probability density functions (conditional) and joint cumulative distribution functions (condi- tional) through several Lemmas and Corollaries. Then, using simulation study, we compare the power of our introduced tests against previously established for detect- ing outliers. Finally, we provide real examples to demonstrate the performance of these tests.

Keywords