BAYESIAN NONPARAMETRIC PREDICTIVE INFERENCE AND BOOTSTRAP TECHNIQUES

We address the question as to whether a prior distribution on the spac e of distribution functions exists which generates the posterior produ ced by Efron's and Rubin's bootstrap techniques, emphasizing the conne ctions with the Dirichlet process. We also introduce a new resampling plan which has two advantages: prior opinions are taken into account a nd the predictive distribution of the future observations is not force d to be concentrated on observed values.