Selection of Largest Multiple Correlation Coefficients: Exact Sample Size Case.
Abstract
In a recent article, Rizvi and Solomon considered the problem of selection of t largest from among k multiple correlation coefficients, each arising from one of k independent p-variate normal distributions with unknown mean vectors and unknown covariance matrices. The problem there is formulated as a ranking problem with a particular choice of an indifference zone in the product parameter space; the main result concerns the minimization of the asymptotic probability of a correct selection for large common sample sizes when the natural selection procedure based on sample multiple correlation coefficients is used for ranking. In view of the limited applicability of the asymptotic solution in the recent article, the present article offers an exact solution to the above problem still using the indifference zone approach and the same natural selection procedure.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 03, 1975
- Accession Number
- ADA012827
Entities
People
- Herbert Solomon
- Khursheed Alam
- M. Haseeb Rizvi
Organizations
- Stanford University