Back To Index Previous Article Next Article Full Text


Statistica Sinica 9(1999), 1131-1140



MULTIPLE HYPOTHESES TESTING WITH

PARTIAL PRIOR INFORMATION



Jian Zhang


Academia Sinica, Beijing and Universite Catholique de Louvain


Abstract: By modifying the optimal criterion of Kudo (1967), an alternative theory is developed for incorporating prior probabilities into multiple hypotheses testing. Unlike Kudo's theory, an explicit optimal test can be obtained. It is shown that the optimal test is a natural extension of some well-known tests such as the maximum likelihood test.



Key words and phrases: Bayesian robust analysis, multiple hypotheses testing, partial prior information.



Back To Index Previous Article Next Article Full Text