BAYESIAN NONPARAMETRIC ESTIMATION OF A BIVARIATE SURVIVAL FUNCTION

Abstract

We use a bivariate reinforced process derived from a Generalized Polya Urn scheme to provide an estimator for a bivariate survival function. The estimator may be considered as a Bayesian nonparametric predictive estimator and it can be obtained easily via an implementation of a Gibbs sampler. Consistency of the estimator is studied.