透過您的圖書館登入
IP:18.119.132.223

摘要


Purpose: In this paper, we discuss the sequential probability ratio test (SPRT) to test the skewness of skew slash distribution with the sample size being a random variable. Design/methodology/approach: The SPRT in general requires an expected number of observations considerably smaller than the fixed number of observations needed by the current most powerful test which controlled the errors of the first and second kinds to the same extent as the sequential test. Findings: The operating characteristic function and average sample number are derived. Simulations are studied to indicate the performances of the proposed test in different scenarios. Practical implications: A real application is discussed to explain the process of the SPRT. Originality/value: The results for the application show that the SPRT only requires fewer observations to draw the same conclusion than the existing fitting test with fixed sample size.

延伸閱讀