Exchangeable and partially exchangeable random partitions

Summary Call a random partition of the positive integerspartially exchangeable if for each finite sequence of positive integersn ₁,...,n _k, the probability that the partition breaks the firstn ₁+...+n_k integers intok particular classes, of sizesn ₁,...,n_k in order of their first elements, has the same valuep(n ₁,...,n_k) for every possible choice of classes subject to the sizes constraint. A random partition is exchangeable iff it is partially exchangeable for a symmetric functionp(n ₁,...n_k). A representation is given for partially exchangeable random partitions which provides a useful variation of Kingman's representation in the exchangeable case. Results are illustrated by the two-parameter generalization of Ewens' partition structure.Research supported by N.S.F. Grants MCS91-07531 and DMS-9404345     

Finite population corrections for ranked set sampling

Ranked set sampling (RSS) for estimating a population mean is studied when sampling is without replacement from a completely general finite populationx=(x ₁,x ₂,...,x _N ). Explicit expressions are obtained for the variance of the RSS estimator and for its precision relative to that of simple random sampling without replacement. The critical term in these expressions involves a quantity =(x–)(x–) where is anN × N matrix whose entries are functions of the population sizeN and the set-sizem, but where does not depend on the population valuesx. A computer program is given to calculate for arbitraryN andm. When the population follows a linear (resp., quadratic) trend, then is a polynomial inN of degree 2m+2 (resp., 2m+4). The coefficients of these polynomials are evaluated to yield explicit expressions for the variance and the relative precision of for these populations. Unlike the case of sampling from an infinite population, here the relative precision depends upon the number of replications of the set sizem.Prepared with partial support from the Statistical Analysis and Computing Branch. Environmental Statistics and Information Division, Office of Policy, Planning, and Evaluation, United States Environmental Protection Agency, Washington, DC under a Cooperative Agreement Number CR-821531. The contents have not been subjected to Agency review and therefore do not necessarily reflect the views of the Agency and no official endorsement should be inferred.     

Absolutely convergent multiple fourier series and multiplicative Lipschitz classes of functions

We consider N-multiple trigonometric series whose complex coefficients c _j1,...,j _N , (j ₁,...,j _N ) ∈ ? ^N , form an absolutely convergent series. Then the series $$ \sum\limits_{(j_1 , \ldots ,j_N ) \in \mathbb{Z}^N } {c_{j_1 , \ldots j_N } } e^{i(j_1 x_1 + \ldots + j_N x_N )} = :f(x_1 , \ldots ,x_N ) $$ converges uniformly in Pringsheim’s sense, and consequently, it is the multiple Fourier series of its sum f, which is continuous on the N-dimensional torus $ \mathbb{T} $ ^N , $ \mathbb{T} $ := [?π, π). We give sufficient conditions in terms of the coefficients in order that >f belong to one of the multiplicative Lipschitz classes Lip (α₁,..., α _N ) and lip (α₁,..., α _N ) for some α₁,..., α _N > 0. These multiplicative Lipschitz classes of functions are defined in terms of the multiple difference operator of first order in each variable. The conditions given by us are not only sufficient, but also necessary for a special subclass of coefficients. Our auxiliary results on the equivalence between the order of magnitude of the rectangular partial sums and that of the rectangular remaining sums of related N-multiple numerical series may be useful in other investigations, too.     

The Embedding Theorem for Cantor Varieties

Let m and n be fixed integers, with 1 m < n. A Cantor variety C _m,n is a variety of algebras with m n-ary and n m-ary basic operations which is defined in a signature ={g₁,...,g_m,f₁,...,f_n} by the identities f_ig₁x₁,...,x_n),...,g_mx₁,...,x_n) = x_i, _i=1,...,n, g_jf₁x₁,...,x_m),...,f_nx₁,...,x_m)) = x_j, _j=1,...,m. We prove the following: (a) every partial C _m,n-algebra A is isomorphically embeddable in the algebra G= A; S(A) of C _m,n; (b) for every finitely presented algebra G= A; S in C _m,n, the word problem is decidable; (c) for finitely presented algebras in _{C m}, the occurrence problem is decidable; (d) _{C m,n} has a hereditarily undecidable elementary theory.     

Trace formulas for a class of subnormal tuples of operators

Trace formulas are established for the product of commutators related to subnormal tuple of operators (S ₁,...,S _n ) with minimal normal extension (N ₁,...,N _n ) satisfying conditions that sp(S _j )/sp(N _j ) is simply-connected with smooth boundary Jordan curve sp(N _i ) and [S _j ^* ,S _j ]^1/2 L ¹,j=1, 2,...,n.Some complete unitary invariants related to the trace formulas are found.This work is supported in part by NSF Grant no. DMS-9101268.     

On Kendall's Process

LetZ₁, …, Z_nbe a random sample of sizen?2 from ad-variate continuous distribution functionH, and letV_i, nstand for the proportion of observationsZ_j,j≠i, such thatZ_j?Z_icomponentwise. The purpose of this paper is to examine the limiting behavior of the empirical distribution functionK_nderived from the (dependent) pseudo-observationsV_i, n. This random quantity is a natural nonparametric estimator ofK, the distribution function of the random variableV=H(Z), whose expectation is an affine transformation of the population version of Kendall's tau in the cased=2. Since the sample version ofτis related in the same way to the mean ofK_n, Genest and Rivest (1993,J. Amer. Statist. Assoc.) suggested that[formula]be referred to as Kendall's process. Weak regularity conditions onKandHare found under which this centered process is asymptotically Gaussian, and an explicit expression for its limiting covariance function is given. These conditions, which are fairly easy to check, are seen to apply to large classes of multivariate distributions.     

Marcinkiewicz multiplier theorem for the Weyl functional calculus

Let {X _j , Y _j , T : 1 ≤ j ≤ n} be a basis satisfying the commutation relation for the Heisenberg Lie algebra . Then we obtain a multi-parameter Marcinkiewicz multiplier theorem for the operator defined by m(X ₁,..., X _n , Y ₁,..., Y _n , T).     

Total Variation Distance for Poisson Subset Numbers

Let n be an integer and A₀,..., A_k random subsets of {1,..., n} of fixed sizes a₀,..., a_k, respectively chosen independently and uniformly. We provide an explicit and easily computable total variation bound between the distance from the random variable , the size of the intersection of the random sets, to a Poisson random variable Z with intensity λ = EW. In particular, the bound tends to zero when λ converges and for all j = 0,..., k, showing that W has an asymptotic Poisson distribution in this regime. Received February 24, 2005     

Real submanifolds in complex spaces

Let(z_(11),..., z_(1N),..., z_(m1),..., z_(mN), w_(11),..., w_(mm)) be the coordinates in C~(mN) +m~2. In this note we prove the analogue of the Theorem of Moser in the case of the real-analytic submanifold M defined as follows W = ZZ~t+ O(3),where W = {w_(ij)}_(1≤i,j≤m)and Z = {z_(ij) }_(1≤i≤m, 1≤j≤N). We prove that M is biholomorphically equivalent to the model W = ZZ~t if and only if is formally equivalent to it.     

On stochastic orders of absolute value of order statistics in symmetric distributions

Let Y₁,…,Y_n be the order statistics of a simple random sample from a finite or infinite population, having median =M. We compare the variables |Y_j−M| and |Y_m−M|, where Y_m is the sample median, that is, for odd n. The comparison is in terms of the likelihood ratio order, which implies stochastic order as well as other orders. The results were motivated by the study of best invariant and minimax estimators for the k/N quantile of a finite population of size N, with a natural loss function of the type , where F_N is the population distribution function, t is an estimate, and g is an increasing function.     

Characterizing optimal sampling of binary contingency tables via the configuration model

A binary contingency table is an m × n array of binary entries with row sums r = (r₁, …, r_m) and column sums c = (c₁, …, c_n). The configuration model generates a contingency table by considering r_i tokens of type 1 for each row i and c_j tokens of type 2 for each column j, and then taking a uniformly random pairing between type‐1 and type‐2 tokens. We give a necessary and sufficient condition so that the probability that the configuration model outputs a binary contingency table remains bounded away from 0 as \begin{align*}N=\sum_{i=1}^m r_i=\sum_{j=1}^n c_j\end{align*} goes to ∞. Our finding shows surprising differences from recent results for binary symmetric contingency tables. © 2012 Wiley Periodicals, Inc. Random Struct. Alg., 2012     

Characterization of Generating Ideals in Some Rings of Entire Functions

Let E be a ring of entire functions on with the operation of pointwise multiplication, and let f ₁,...,f _m be a set of nonzero elements in E. The ideal E(f ₁,...,f _m ) in E with generators f ₁,...,f _m is said to be generating if E(f ₁,...,f _m ) = E. The generating ideals in rings of entire functions on determined by the growth of their maximum moduli are characterized in terms of the distribution of the zero sets of their generators. Under the additional condition of rapid variation of the weight sequences determining the ring, criteria for generating ideals are established; they are stated in terms of d(z) max _{1 j m} d _j (z), where d _j (z) is the distance from a point to the zero set of f _j for 1 j m. It is shown that, in rings of entire functions of finite or minimal type with respect to a given order, a similar characterization (i.e., in terms of d(z)) cannot be given.     

Limit Laws for Sums of Independent Random Products: the Lattice Case

Let {V _i,j ;(i,j)∈ℕ²} be a two-dimensional array of independent and identically distributed random variables. The limit laws of the sum of independent random products

Spheres in Hermitian Symmetric Spaces and Flag Manifolds

This paper contains a proof of the following property of compact irreducible Hermitian symmetric spaces. If H=G/K where G is a compact simply connected simple Lie group, T a maximal torus of G and F(T,H)=|E ₁,...,E _m is the fixed point set of T on H, then for each pair E _i , E _j there is a 2-dimentional sphere N _ij H such that E _i and E _j are antipodal points of N _ij.     

Berry–Esseen Bound for a Sample Sum from a Finite Set of Independent Random Variables

Let {X ₁,...,X _N} be a set of N independent random variables, and let S _n be a sum of n random variables chosen without replacement from the set {X ₁,...,X _N} with equal probabilities. In this paper we give an estimate of the remainder term for the normal approximation of S _n under mild conditions.     

Proof of the Boltzmann-Sinai ergodic hypothesis for typical hard disk systems

We consider the system of N (2) hard disks of masses m₁,...,m_N and radius r in the flat unit torus ². We prove the ergodicity (actually, the B-mixing property) of such systems for almost every selection (m₁,...,m_N;r) of the outer geometric parameters.Dedicated to Ya. G. Sinai honoring his 65th birthdayMathematics Subject Classification (2000) 37D50, 34D05     

Coherent random allocations, and the Ewens-Pitman formula

Assume that there is a random number K of positive integer random variables S₁, …, S_K that are conditionally independent given K and all have identical distributions. A random integer partition N = S₁ + S₂ + … + S_K arises, and we denote by P_N the conditional distribution of this partition for a fixed value of N. We prove that the distributions {P_N} _N=1 ^∞ form a partition structure in the sense of Kingman if and only if they are governed by the Ewens-Pitman Formula. The latter generalizes the celebrated Ewens sampling formula, which has numerous applications in pure and applied mathematics. The distributions of the random variables K and S_j belong to a family of integer distributions with two real parameters, which we call quasi-binomial. Hence every Ewens-Pitman distribution arises as a result of a two-stage random procedure based on this simple class of integer distributions. Bibliography: 25 titles. This paper is an edited and actualized version of the unpublished PDMI preprint 21/1995. Further development of the ideas of this work can be found in [21, 25]. A number of detected misprints was fixed without notice, the bibliography was extended beyond the original 19 references, and a few comments were added as footnotes. (Comments by Alexander Gnedin.) __________ Translated from Zapiski Nauchnykh Seminarov POMI, Vol. 325, 2005, pp. 127–145.     

A form of multivariate gamma distribution

Let V _i, i=1,..., k, be independent gamma random variables with shape i, scale , and location parameter i, and consider the partial sums Z ₁=V ₁, Z ₂=V ₁+V ₂,..., Z _k=V ₁+...+V _k. When the scale parameters are all equal, each partial sum is again distributed as gamma, and hence the joint distribution of the partial sums may be called a multivariate gamma. This distribution, whose marginals are positively correlated has several interesting properties and has potential applications in stochastic processes and reliability. In this paper we study this distribution as a multivariate extension of the three-parameter gamma and give several properties that relate to ratios and conditional distributions of partial sums. The general density, as well as special cases are considered.     

On the Volume of Flowers in Space Forms

Let f(x ₁,..., x _N ) be a lattice polynomial in N variables, in which each variable occurs exactly once, B ₁,..., B _N be smoothly moving balls in the hyperbolic, Euclidean, or spherical space. Introducing a suitable modification of the Dirichlet–Voronoi decomposition, we prove a formula for the derivative of the volume of the domain f(B ₁,..., B _N ). As an application of the formula, we show that the volume increases if the balls move continuously in such a way that the functions _ij d _ij increase for all 1 i < j N, where _ij is a sign depending on f, d _ij is the distance between the centers of B _I and B _j .     

A generalization of the inversion formulas of systems of power series in systems of implicit functions

Using a multidimensional analog of the logarithmic residue, equations are derived expressing the coefficients of the power series of implicit functionsx _j =_j(w)=_j(w₁,...,w_m), j=1,...,n, defined by the system of equations f_j(w, x)=F_j (w₁,...,w_m:z₁,...,x _n )=0, j=1,...,n,f _j , (0, 0)=0, F_j(0, 0)/z_k=_jk in a neighborhood of the point (0, 0)C _(w,x) ^m+n , in terms of the coefficients of the power series of the functions F_j(w, z), j=1, ..., n. As a corollary, well-known formulas are obtained for the inversion of multiple power series.Translated from Matematicheskie Zametki, Vol. 23, No. 1, pp. 47–54, January, 1978.