For costly and/or destructive tests, the sequential method with a proper maximum sample size is needed. Based on Koopman-Darmois distributions, this paper proposes the method of sequential mesh, which has an acceptable maximum sample size. In comparison with the popular truncated sequential probability ratio test, our method has the advantage of a smaller maximum sample size and is especially applicable for costly and/or destructive tests.
method of sequential mesh sequential probability ratio test (SPRT) Koopman-Darmois distributions maximum sample size
This is a preview of subscription content, log in to check access.
Ingeborg T, Paulus A H N. Early stopping in clinical trials and epidemiologic studies for “futility”: conditional power versus sequential analysis. J Clinical Epidemiology, 2003, 56: 610–617CrossRefGoogle Scholar
International Electrotechnical Commission. International Standard of IEC 1123: Reliability testing compliance test plans for success ratio, 1991–12Google Scholar
Lorden G. 2-SPRT’s and the modified Kiefer-Weiss problem of minimizing an expected sample size. Ann Math Statist, 1976, 4: 281–291zbMATHMathSciNetGoogle Scholar
Marano S, Willett P, Matta V. Sequential testing of sorted and transformed data as an efficient way to implement long GLRTs. IEEE Trans Signal Process, 2003, 51: 325–337CrossRefMathSciNetGoogle Scholar
Pu X L, Yan Z G, Mao S S, et al. The sequential mesh test for a proportion (in Chinese). J East China Norm Univ Natur Sci Ed, 2006, 1: 63–71MathSciNetGoogle Scholar