Sequential Analysis for Testing the Equality of Two Distributions Based on Early Failed Ranked Set Samples

• Activity Number: 303 - Recent Developments on Order-Related Designs and Inferences
• Type: Topic Contributed
• Date/Time: Wednesday, August 5, 2020 : 10:00 AM to 11:50 AM
• Sponsor: Korean International Statistical Society
• Author(s): Tao Li*
• Companies: Shanghai University of Fianance and Economics
• Keywords: precedence statistics; sequential test; life testing; Lehmann alternative; two-sample problem

Abstract

In this paper, we describe the implantation of ranked set sampling scheme into sequential life-time testing experiments, establish the test statistics. The null distribution is derived, several ?-spending sequences are considered, and the corresponding critical values for each ?-spending sequence are presented. Moreover, the power and the stopping time under both Lehmann alternative and location-shift alternative are discussed. Through a comparison of the exact expected stopping locations under the Lehmann alternative as well as the simulated expected stopping locations under the location-shift alternative, we observe that the proposed sequential precedence tests are more efficient than the nonsequential precedence test in terms of saving experimental units. Furthermore, the impact of ranking errors are also presented.