Approximating de Finetti's measures for partially exchangeable sequences (Q1573130)

The authors show how to approximate de Finetti's measure of a partially exchangeable sequence by a mixture of products of Dirichlet measures, explicitly built once the approximation error has been fixed. They claim that this method, when it is meaningfully applied, is an ``objectivist'' procedure for the assessment of prior distributions, since it essentially uses available data (relative to phenomena similar to that they consider) to construct the prior.