Random weighted averages, partition structures and generalized arcsine laws

From MaRDI portal
Publication:6300623




Abstract: This article offers a simplified approach to the distribution theory of randomly weighted averages or P-means MP(X):=sumjXjPj, for a sequence of i.i.d.random variables X,X1,X2,ldots, and independent random weights P:=(Pj) with Pjge0 and sumjPj=1. The collection of distributions of MP(X), indexed by distributions of X, is shown to encode Kingman's partition structure derived from P. For instance, if Xp has Bernoulli(p) distribution on 0,1, the nth moment of MP(Xp) is a polynomial function of p which equals the probability generating function of the number Kn of distinct values in a sample of size n from P: E(MP(Xp))n=EpKn. This elementary identity illustrates a general moment formula for P-means in terms of the partition structure associated with random samples from P, first developed by Diaconis and Kemperman (1996) and Kerov (1998) in terms of random permutations. As shown by Tsilevich (1997) if the partition probabilities factorize in a way characteristic of the generalized Ewens sampling formula with two parameters (alpha,heta), found by Pitman (1992), then the moment formula yields the Cauchy-Stieltjes transform of an (alpha,heta) mean. The analysis of these random means includes the characterization of (0,heta)-means, known as Dirichlet means, due to Von Neumann (1941), Watson (1956) and Cifarelli and Regazzini (1990) and generalizations of L'evy's arcsine law for the time spent positive by a Brownian motion, due to Darling (1949) Lamperti (1958) and Barlow, Pitman and Yor (1989).











This page was built for publication: Random weighted averages, partition structures and generalized arcsine laws

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6300623)