On comparing binomial probabilities from a bayesian viewpoint

Abstract
A usable procedure is presented which orders binomial probabilities. An essentially Bayesian approach to the multiple comparisons problem is applied to the binomial model, and the exact nature of the orderings is given. This approach allows the utilization of prior information (in the form of beta densities) on the individual probabilities, possibly unequal sample sizes, and both constant and linear loss functions.

This publication has 3 references indexed in Scilit: