Partition structures, Polya urns, the Ewens sampling formula, and the ages of alleles (Q1086510)

This paper has two main aims. The first is to explain the connection between various recent results on partition structures, Polya urns, the Ewens sampling formula, and the ages of alleles, and as a result to unify a number of seemingly diverse approaches. Further, the sampling formula is characterized as being the only partition structure which may be generated by an urn-like mechanism. The second purpose of this paper is to advertise a flexibility of approach. Faced with a question concerning the age ordering of alleles, this paper has outlined a number of equivalent ways of thinking about this question, and it is hard to envisage a problem which will not yield easily to at least one of these approaches, the consequences of reversibility, or genealogical arguments. In particular, quantities associated with the Poisson-Dirichlet distribution now can be obtained readily from at least one of the other approaches.