On the Evidence Needed to Reach Agreed Action between Adversaries, with Application to Acceptance Sampling

摘要: Abstract Two decision makers disagree about a quantity of interest to them both. One them, the “consumer,” has choice two decisions that are affected by quantity. The other, “manufacturer,” offers perform an agreed type experiment it is hoped will change consumer's view and hence decision. This article devoted evaluation how much experimentation should be done. Binomial, Poisson, normal likelihoods, together with their conjugate utilities probabilities, considered illustrated numerical cases. scenario here arises in applications quality control, bidding, drug testing, marketing, sales.