A Restricted Subset Selection Approach to Ranking and Selection Problems.
Abstract
The goal in the paper is to study single sample procedures which give more flexibility to the experimenter than does either the fixed subset size rule or the subset selection procedure by allowing him to specify an upper bound, m, on the number of populations included in the selected subset. Should the data clearly indicate that a particular population is best, this type of rule retains the advantage of the subset selection procedure over the fixed size subset rule in allowing selection of fewer than m populations. On the other hand if the data make the choice of the best population less obvious, this rule selects a larger subset for further study but guarantees that no more than m populations are selected. Such procedures will be called restricted subset selection procedures.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1973
- Accession Number
- AD0766547
Entities
People
- Thomas J. Santner
Organizations
- Purdue University