Optimal sample size allocation for integrated test scheme

Abstract

Integrated test scheme (ITS) make comprehensive use of simulation and field test to evaluate system performance, which takes on merits of low cost, full-scale assessment, and high precision. Sample size determination (SSD) methods for ITS are analyzed, which reveals the problem and contradiction in classical method and Bayesian method using standard power prior for design. Design effect of experiment is proposed considering the influence of prior sample size, which is used for elicitation of modified power prior for SSD of ITS. Taking average posterior variance for parameter of interest as the output precision, optimization equation for SSD of ITS under required posterior precision constraint is designed, whose effectiveness is illustrated with a utilization example.