Probability Density When an Independent Observation is from an Estimated Multivariate Normal Population.
Abstract
The purpose of this paper is to derive the exact probability density for a sample vector from a multivariate normal population with estimated parameters. A general theorem and two applications-oriented corollaries were developed toward this end. When their assumptions can be met, these corollaries are useful for answering questions such as: 'How unlikely is it that a particular observation arose from a specified population' and 'To which of several populations does a particular observation most likely belong.' Some methods for applying the results are discussed, with emphasis on answering these two questions. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 15, 1971
- Accession Number
- AD0726697
Entities
People
- A. C. Bittner Jr.