Multiple integral representation for functionals of Dirichlet processes

Abstract: We point out that a proper use of the Hoeffding--ANOVA decomposition for symmetric statistics of finite urn sequences, previously introduced by the author, yields a decomposition of the space of square-integrable functionals of a Dirichlet--Ferguson process, written

L^{2} (D)

, into orthogonal subspaces of multiple integrals of increasing order. This gives an isomorphism between

L^{2} (D)

and an appropriate Fock space over a class of deterministic functions. By means of a well-known result due to Blackwell and MacQueen, we show that each element of the

n

th orthogonal space of multiple integrals can be represented as the

L^{2}

limit of

U

-statistics with degenerate kernel of degree

n

. General formulae for the decomposition of a given functional are provided in terms of linear combinations of conditioned expectations whose coefficients are explicitly computed. We show that, in simple cases, multiple integrals have a natural representation in terms of Jacobi polynomials. Several connections are established, in particular with Bayesian decision problems, and with some classic formulae concerning the transition densities of multiallele diffusion models, due to Littler and Fackerell, and Griffiths. Our results may also be used to calculate the best approximation of elements of

L^{2} (D)

by means of

U

-statistics of finite vectors of exchangeable observations.

Review text

The main purpose of this article is a chaotic representation for square integrable functionals of a Dirichlet process, very analogous to the root usual Wiener or Poisson chaos, except that no regularity of the functional is needed here. The Dirichlet (or Dirichlet-Ferguson) process is a random probability measure \(D\), defined over a measured space \((A,C\ell,\alpha)\), \(\alpha\) a given finite Borel measure, by the requirement that for any finite measurable partition of \(A:(C_1,\dots, C_n)\) the vector \((D(C_1),\dots, D(C_n))\) has the Dirichlet distribution with parameters \((\alpha(C_1),\dots,\alpha(C_n))\). It is known that such Dirichlet process can be got as a.s. limit of infinite urn Pólya scheme \((X_n,n\geq 1)\), \(\alpha\) representing the initial composition of the urn. Then \((X_n)\) is an infinite exchangeable sequence, which is i.i.d. with common law \(D\), conditionally on \(D\). The Fock space associated with \((X_n)\) is known in statistics as Hoeffding space. The chaotic representation for \(L^2(D)\) induces (as for the Wiener chaos) a Hilbert space isomorphism with this Hoeffding space. The author provides finally an application of this chaotic representation, to Bayesian nonparametric \(U\)-statistics.

Recommendations

Cites work

Cited in

(13)

This page was built for publication: Multiple integral representation for functionals of Dirichlet processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1002577)