On Two-Stage Prodcedures for Finding A Population Better than a Control.
Abstract
A natural class of two-stage procedures is proposed which can be completely described and studied in terms of Neyman-Pearson testing theory, where the unsymmetry of tests, however, can be overcome to a considerable extent. As a typical result it is shown that optimality of tests carries over to optimality of two-stage procedures. Finally, under normality, comparisons are made in case of k = 2 with certain Bayesian procedures.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1980
- Accession Number
- ADA100950
Entities
People
- Klaus J. Miescke
Organizations
- Purdue University