On the Uniform Consistency of Bayes Estimates for Multinomial Probabilities
Diaconis, P. ; Freedman, D.
Ann. Statist., Tome 18 (1990) no. 1, p. 1317-1327 / Harvested from Project Euclid
A $k$-sided die is thrown $n$ times, to estimate the probabilities $\theta_1, \ldots, \theta_k$ of landing on the various sides. The MLE of $\theta$ is the vector of empirical proportions $p = (p_1, \ldots, p_k)$. Consider a set of Bayesians that put uniformly positive prior mass on all reasonable subsets of the parameter space. Their posterior distributions will be uniformly concentrated near $p$. Sharp bounds are given, using entropy. These bounds apply to all sample sequences: There are no exceptional null sets.
Publié le : 1990-09-14
Classification:  Bayes estimates,  consistency,  Laplace's method,  multinomial,  Bernstein-von Mises theorem,  62A15,  62E20
@article{1176347751,
     author = {Diaconis, P. and Freedman, D.},
     title = {On the Uniform Consistency of Bayes Estimates for Multinomial Probabilities},
     journal = {Ann. Statist.},
     volume = {18},
     number = {1},
     year = {1990},
     pages = { 1317-1327},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176347751}
}
Diaconis, P.; Freedman, D. On the Uniform Consistency of Bayes Estimates for Multinomial Probabilities. Ann. Statist., Tome 18 (1990) no. 1, pp.  1317-1327. http://gdmltest.u-ga.fr/item/1176347751/