Title: “Bayesian Hierarchical Mixture Modelling”.
Abstract: “Bayesian hierarchical models provide a useful model structure for combining information from related experiments. In their basic formulation, it is assumed that the parameters of the model satisfy the assumption of exchangeability. We investigate a generalization of the basic Bayesian hierarchical model by assuming a partial exchangeability structure
which is modelled with a finite mixture of normals with unknown number of components. We will discuss why this assumption is more appealing and provide illustrative examples from finance and genetics.