Creator |
|
Language |
|
Publisher |
|
|
Date |
|
Source Title |
|
Vol |
|
Issue |
|
First Page |
|
Last Page |
|
Publication Type |
|
Access Rights |
|
Crossref DOI |
|
Related DOI |
|
|
Related URI |
|
|
Relation |
|
|
Abstract |
When independent random variables $ X_i $ $ ( i = 1, 2, cdots, k ) $ have probability density $ f_{\theta_i} $, with monotone likelihood ratio respectively, we consider testing $ H_0 : mathrm{min} (\t...heta_1, \theta_2, cdots, \theta_k) = \theta^ast $ vs. $ H_1 : mathrm{min} (\theta_1, \theta_2, cdots, \theta_k) > \theta^ast $ for a constant $ \theta^ast $. We give a class of similar tests and find an unbiased test in this class. We apply and extend the arrangement ordering arguments. It is also proved that this unbiased test has a monotone power function. A modification to testing $ H_0^ast : mathrm{min} (\theta_1, \theta_2, cdots, \theta_k) leqq \theta^ast $ vs. $ H_1 $ is also considered.show more
|