Reference priors based on composite likelihoods

In this paper we propose reference priors obtained by maximizing the average adivergence from the posterior distribution, when the latter is computed using a composite likelihood. Composite posterior distributions have already been considered in [7] and [8], when a full likelihood for the data is too complex or even not available. The use of a curvature corrected composite posterior distribution, as in [8], allows to apply the method in [6] for maximizing the asymptotic Bayes risk associated to an adivergence. The result is a Jeffreys type prior that is proportional to the square root of the determinant of the Godambe information matrix.