On Selection Rules for Finite Mixtures of Distributions,
Abstract
A class of consistent selection procedures is proposed for the selection of the t largest proportions of finite mixture of distributions. The approach is based on the indifference zone formulation. Some asymptotic optimal properties are shown to hold. For special case of m = 2, the number of components in each of k mixture populations, some results of the asymptotic relative efficiency are obtained. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1977
- Accession Number
- ADA045106
Entities
People
- Shanti Gupta
- Wen-tao Huang
Organizations
- Purdue University