Objective priors for the bivariate normal model
Berger, James O. ; Sun, Dongchu
Ann. Statist., Tome 36 (2008) no. 1, p. 963-982 / Harvested from Project Euclid
Study of the bivariate normal distribution raises the full range of issues involving objective Bayesian inference, including the different types of objective priors (e.g., Jeffreys, invariant, reference, matching), the different modes of inference (e.g., Bayesian, frequentist, fiducial) and the criteria involved in deciding on optimal objective priors (e.g., ease of computation, frequentist performance, marginalization paradoxes). Summary recommendations as to optimal objective priors are made for a variety of inferences involving the bivariate normal distribution. ¶ In the course of the investigation, a variety of surprising results were found, including the availability of objective priors that yield exact frequentist inferences for many functions of the bivariate normal parameters, including the correlation coefficient.
Publié le : 2008-04-15
Classification:  Reference priors,  matching priors,  Jeffreys priors,  right-Haar prior,  fiducial inference,  frequentist coverage,  marginalization paradox,  rejection sampling,  constructive posterior distributions,  62F10,  62F15,  62F25,  62A01,  62E15,  62H10,  62H20
@article{1205420525,
     author = {Berger, James O. and Sun, Dongchu},
     title = {Objective priors for the bivariate normal model},
     journal = {Ann. Statist.},
     volume = {36},
     number = {1},
     year = {2008},
     pages = { 963-982},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1205420525}
}
Berger, James O.; Sun, Dongchu. Objective priors for the bivariate normal model. Ann. Statist., Tome 36 (2008) no. 1, pp.  963-982. http://gdmltest.u-ga.fr/item/1205420525/