The use of correlated information on treatment effects when selecting the best treatment

Abstract

The selection of the best variety of a crop to grow in a particular region can be based solely on the yields of the varieties in trials within that region or on the yields of the varieties averaged over trials within the region and trials in a number of similar regions. For a wide range of values of the parameters, the better of these two methods is shown to be almost as effective as selection based on an index combining, in an optimal way, the information from the particular region and from the other regions. A similar situation in the choice of a medical treatment for a particular type of patient is mentioned.