Abstract : This paper investigates the nonasymptotic properties of Bayes procedures for estimating an unknown distribution from nn i.i.d. observations. We assume that the prior is supported by a model (S,h)(S,h) (where hh denotes the Hellinger distance) with suitable metric properties involving the number of small balls that are needed to cover larger ones. We also require that the prior put enough probability on small balls.
We consider two different situations. The simplest case is the one of a parametric model containing the target density for which we show that the posterior concentrates around the true distribution at rate View the MathML source1/n. In the general situation, we relax the parametric assumption and take into account a possible misspecification of the model. Provided that the Kullback–Leibler Information between the true distribution and the model SS is finite, we establish risk bounds for the Bayes estimators.