On Isotonic Selection Rules for Binomial Populations Better than a Standard.
Abstract
The problem of selecting populations better than a control with respect to a location parameter under ordering prior has been considered. In this paper we consider the case of binomial populations (important for discrete data) for which the parameters of interest are not the location parameters as was the case studied in 4. We consider both cases when the parameter of control is known and unknown. For the case of known control, we propose an isotonic procedure which is given. The results in this section deal with both cases, namely, the sample sizes all equal, and unequal. Some recursive relations are derived for computing the constants required for the proposed procedure. When the control is unknown, a conditional isotonic procedure is proposed. This procedure provides a conservative solution for the unconditional procedure. Brief tables of associated constants for the proposed procedures are given.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1982
- Accession Number
- ADA118548
Entities
People
- Shanti Gupta
- Wen-tao Huang
Organizations
- Purdue University