EXACT BAYESIAN TESTS OF SHARP HYPOTHESES.
SYSTEM DEVELOPMENT CORP SANTA MONICA CALIF
Pagination or Media Count:
The Bayesian theory for testing a hypothesis defined by fixed values of parameters is presented in general terms without approximations. Arbitrary positive prior probability is attached to the hypothesis. The conditional distribution of the nuisance parameters given the parameters defining the hypothesis is assumed to be continuous in the latter at their values for the hypothesis. Families of integrable conjugate prior distributions are assigned under the alternative hypothesis for the problem of equality of Bernoulli parameters, for multivariate extensions, and for multivariate normal problems. Author
- Statistics and Probability