Locally Optimal Tests in Cardioid Distribution

碩士 === 國立中興大學 === 應用數學系 === 87 === We are adopting the concept of Locally Most Powerful test and Locally Most Powerful Invariate test to derive optimal tests with respect to one-sided for one unknown parameter and two-sided for two unknown parameters probl...

Full description

Bibliographic Details
Main Authors: C-J Tsai, 蔡嘉仁
Other Authors: H-J Chang
Format: Others
Language:en_US
Published: 1999
Online Access:http://ndltd.ncl.edu.tw/handle/55528025062943018021
Description
Summary:碩士 === 國立中興大學 === 應用數學系 === 87 === We are adopting the concept of Locally Most Powerful test and Locally Most Powerful Invariate test to derive optimal tests with respect to one-sided for one unknown parameter and two-sided for two unknown parameters problems,respectively. In this thesis, the problem under discussion is that whether a circular random sample is from the uniform distribution or from Cardioid distribution. After these two respective optimal tests have been developed, some optimal properties pertaining to these tests are established, as well. Finally, the Monte-Carlo simulation is carried out to obtain the numerical result of the theoretical consequence, and the performance of simulation is far better than what we have expected.