Accession Number:

ADA025452

Title:

Exact and Approximate Bayesian Solutions for Inference about Variance Components and Multivariate Inadmissibility.

Descriptive Note:

Final technical rept. 1 May 72-1 Jun 76,

Corporate Author:

MICHIGAN UNIV ANN ARBOR DEPT OF STATISTICS

Personal Author(s):

• Hill,Bruce M.

Report Date:

1976-02-01

Pagination or Media Count:

37.0

Abstract:

Exact and approximate Bayesian solutions for inference about variance components and multivariate inadmissibility are obtained, using a class of prior distributions which allows realistic forms of prior knowledge to be incorporated. Results are based upon new mathematical representations for the Appell and ordinary hypergeometric functions. Author

Descriptors:

• *MULTIVARIATE ANALYSIS
• PROBABILITY DISTRIBUTION FUNCTIONS
• RANDOM VARIABLES
• ANALYSIS OF VARIANCE
• BAYES THEOREM
• INTEGRAL TRANSFORMS
• HYPERGEOMETRIC FUNCTIONS

Subject Categories:

• Statistics and Probability

Distribution Statement:

APPROVED FOR PUBLIC RELEASE