On the gain in power when using additional observations |
| |
Authors: | Olaf Krafft Ludger Rtischendorf |
| |
Institution: | Technical University of Aachen , Aachen, West Germany |
| |
Abstract: | In the present paper the question in which cases additional observations are useless for testing two simple hypotheses is considered. It turns out that in all but trivial cases two additional observations strictly decrease the risk, while for one additional observation a characterization of this problem is given. |
| |
Keywords: | Bayes risk testing simple |
|