Anzeige
Erschienen in:
2018 | OriginalPaper | Buchkapitel
Prior Information in Bayesian Linear Multivariate Regression
verfasst von : J. Casaca
Erschienen in: Modeling, Dynamics, Optimization and Bioeconomics III
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
The paper introduces the Bayesian approach to multivariate regression analysis, from a subjective point of view. A review of non-informative and informative priors adequate to practical situations is carried out. The marginal posteriors of the regression coefficients and the variance factors corresponding to the Laplace, Jeffreys and conjugate priors, as well as the respective modes, are presented. Of note is the fact that Laplace and Jeffreys priors, as it would be expected of non-informative priors, yield maximum posterior estimates of the regression coefficients identical to the maximum likelihood estimate.