ON SOME SELECTION AND RANKING PROCEDURES WITH APPLICATIONS TO MULTIVARIATE POPULATIONS.
Abstract
A problem of subset selection for parameters which are not necessarily scale or location parameters is considered. A general theorem dealing with the infimum of the probability of a correct selection for parameters occurring in densities which are Poisson mixtures of arbitrary densities on the interval (0, infinity) is proved. This theorem is applied to obtain the minimum value of the probability of a correct selection in several cases where multivariate normal populations are ranked according to a given scalar quantity.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1965
- Accession Number
- AD0623217
Entities
People
- Shanti Gupta
- William J. Studden
Organizations
- Purdue University