Double stepwise likelihood ratio test for one-sided composite hypotheses

In this paper, we propose a novel sequential test for one-sided composite hypotheses based on a stepwise likelihood ratio statistics. Its stopping boundaries are constant and can be easily determined by our provided searching algorithm. The theoretical results in this paper show that our proposed test satisfies two type error constraints and has a finite stopping time. The simulation studies show that our proposed test combines the advantages of the SPRT and the 2-SPRT and minimizes expected sample size in pointwise. To illustrate the proposed test, a real example is analyzed.