Testing the Equivalence of Binomial and Poisson Distributions

• Main Authors: Yen-Tzu Chen, 陳妍孜
• Other Authors: Miin-Jye Wen
• Format: Others
• Language: en_US
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/73194715352144189180

碩士 === 國立成功大學 === 統計學系碩博士班 === 95 === A range test is proposed for testing the hypothesis of equivalence of binomial proportions and Poisson parameters, respectively against the alternative hypothesis of inequivalence. Variance stabilization transformations for both binomial and Poisson random variables are used to remove unknown parameters in the variances. Both the maximum level and the minimum power of the proposed test associated with these hypotheses can be approximated by normal theory at the corresponding least favorable configurations of parameters. For a given level and a given power, the critical value and the required sample size for an experiment can be simultaneously determined, and a computer program is developed for such purpose.