Bayesian estimation of the dispersion matrix of a multivariate normal distribution
- 1 January 1985
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics - Theory and Methods
- Vol. 14 (5) , 1019-1034
- https://doi.org/10.1080/03610928508828960
Abstract
The estimation of the dispersion matrix of a multivariate normal distribution with zero mean on the basis of a random sample is discussed from a Bayesian view. An inverted-Wishart distribu- tion for the dispersion is taken, with its defining matrix of intraclass form. Some consistency properties are described. The posterior distribution is found and its mode investigated as a possible estimate in preference to that of maximum likelihoodKeywords
This publication has 4 references indexed in Scilit:
- The Role of Exchangeability in InferenceThe Annals of Statistics, 1981
- Bayesianism: Its Unifying Role for Both the Foundations and Applications of StatisticsInternational Statistical Review, 1974
- Bayes Estimates for the Linear ModelJournal of the Royal Statistical Society Series B: Statistical Methodology, 1972
- Ridge Regression: Biased Estimation for Nonorthogonal ProblemsTechnometrics, 1970