Ranges of Posterior Measures for Priors with Unimodal Contaminations
Sivaganesan, S. ; Berger, James O.
Ann. Statist., Tome 17 (1989) no. 1, p. 868-889 / Harvested from Project Euclid
We consider the problem of robustness or sensitivity of given Bayesian posterior criteria to specification of the prior distribution. Criteria considered include the posterior mean, variance and probability of a set (for credible regions and hypothesis testing). Uncertainty in an elicited prior, $\pi_0$, is modelled by an $\varepsilon$-contamination class $\Gamma = \{\pi = (1 - \varepsilon)\pi_0 + \varepsilon q, q \in Q\}$, where $\varepsilon$ reflects the amount of probabilistic uncertainty in $\pi_0$, and $Q$ is a class of allowable contaminations. For $Q = \{$all unimodal distributions$\}$ and $Q = \{\text{all symmetric unimodal distributions}\}$, we determine the ranges of the various posterior criteria as $\pi$ varies over $\Gamma$.
Publié le : 1989-06-14
Classification:  Bayesian robustness,  $\varepsilon$-contamination classes of priors,  posterior mean,  posterior variance,  posterior probability,  62F15,  62F35,  62C10
@article{1176347148,
     author = {Sivaganesan, S. and Berger, James O.},
     title = {Ranges of Posterior Measures for Priors with Unimodal Contaminations},
     journal = {Ann. Statist.},
     volume = {17},
     number = {1},
     year = {1989},
     pages = { 868-889},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176347148}
}
Sivaganesan, S.; Berger, James O. Ranges of Posterior Measures for Priors with Unimodal Contaminations. Ann. Statist., Tome 17 (1989) no. 1, pp.  868-889. http://gdmltest.u-ga.fr/item/1176347148/