Abstract
A sequential Bayes decision procedure for a binomial parameter is developed. It is assumed that there are three terminal decisions and that a prior distribution of the parameter belongs to the beta family. The decision boundary is obtained numerically by introducing losses, and proceeding at each stage by maximizing the expected gain. The idea of superimposing two decision boundaries is proposed.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have