Abstract: An optimal (Bayes)
sequential plan minimizes the (Bayes) risk function which takes into account the
decision loss, observation (variable) cost, and group (fixed) cost. In general,
determining the optimal sequential plan remains an open problem mainly because
it requires risk-optimization over a rather unstructured set of all plans. An -Bayes sequential plan is a
sequential plan comparable to the Bayes plan yet in general more computationally
feasible. Methods of deriving upper and lower bounds for the Bayes sequential
plan are shown which are then used to define an e-Bayes
sequential plan.
Keywords and phrases: -Bayes, risk function,
sequential plan.