On Minimizing and Stationary Sequences of a New Class of Merit Functions for Nonlinear Complementarity Problems

Motivated by the work of Fukushima and Pang (Ref. 1), we study the equivalent relationship between minimizing and stationary sequences of a new class of merit functions for nonlinear complementarity problems (NCP). These merit functions generalize that obtained via the squared Fischer–Burmeister NCP function, which was used in Ref. 1. We show that a stationary sequence {x^k} /Reⁿ is a minimizing sequence under the condition that the function value sequence {F(x ^k)} is bounded above or the Jacobian matrix sequence {F(x ^k)} is bounded, where F is the function involved in NCP. The latter condition is also assumed by Fukushima and Pang. The converse is true under the assumption of {F(x ^k)} bounded. As an example shows, even for a bounded function F, the boundedness of the sequence {F(x ^k)} is necessary for a minimizing sequence to be a stationary sequence.     

Transforming binary sequences using priority queues

A priority queue transforms an input sequence into an output sequence which is a re-ordering of the sequence . The setR of all such related pairs is studied in the case that is a binary sequence. It is proved thatR is a partial order and that ¦R¦=c _n+1, the (n+1)th Catalan number. An efficient (O(n ²)) algorithm is given for computing the number of outputs achievable from a given input.     

Exact Distribution of the Local Score for Markovian Sequences

Let be a sequence of letters taken in a finite alphabet Θ. Let be a scoring function and the corresponding score sequence where X _i = s(A _i ). The local score is defined as follows: . We provide the exact distribution of the local score in random sequences in several models. We will first consider a Markov model on the score sequence , and then on the letter sequence . The exact P-value of the local score obtained with both models are compared thanks to several datasets. They are also compared with previous results using the independent model.     

An Example of a Positive Semidefinite Double Sequence Which is Not a Moment Sequence

The first explicit example of a positive semidefinite double sequence which is not a moment sequence was given by Friedrich. We present an example with a simpler definition and more moderate growth as (m, n) .     

Syndetically Hypercyclic Operators

Given a continuous linear operator T L(x) defined on a separable -space X, we will show that T satisfies the Hypercyclicity Criterion if and only if for any strictly increasing sequence of positive integers such that the sequence is hypercyclic. In contrast we will also prove that, for any hypercyclic vector x X of T, there exists a strictly increasing sequence such that and is somewhere dense, but not dense in X. That is, T and do not share the same hypercyclic vectors.     

On Powers of Stieltjes Moment Sequences,I

For a Bernstein function f the sequence s_n=f(1)·...· f(n) is a Stieltjes moment sequence with the property that all powers s_n^c,c>0 are again Stieltjes moment sequences. We prove that is Stieltjes determinate for c≤ 2, but it can be indeterminate for c>2 as is shown by the moment sequence , corresponding to the Bernstein function f(s)=s. Nevertheless there always exists a unique product convolution semigroup such that ρ_c has moments . We apply the indeterminacy of for c>2 to prove that the distribution of the product of p independent identically distributed normal random variables is indeterminate if and only if p≥ 3     

On the projected subgradient method for nonsmooth convex optimization in a Hilbert space

We consider the method for constrained convex optimization in a Hilbert space, consisting of a step in the direction opposite to an _k -subgradient of the objective at a current iterate, followed by an orthogonal projection onto the feasible set. The normalized stepsizes _k are exogenously given, satisfying _{k=0 k} = , _{k=0 k} ² < , and _k is chosen so that _{k k} for some > 0. We prove that the sequence generated in this way is weakly convergent to a minimizer if the problem has solutions, and is unbounded otherwise. Among the features of our convergence analysis, we mention that it covers the nonsmooth case, in the sense that we make no assumption of differentiability off, and much less of Lipschitz continuity of its gradient. Also, we prove weak convergence of the whole sequence, rather than just boundedness of the sequence and optimality of its weak accumulation points, thus improving over all previously known convergence results. We present also convergence rate results. © 1998 The Mathematical Programming Society, Inc. Published by Elsevier Science B.V.Research of this author was partially supported by CNPq grant nos. 301280/86 and 300734/95-6.     

Convergence of intermediate rows of minimal polynomial and reduced rank extrapolation tables

Let {x _m} _m ^=0/ be a vector sequence obtained from a linear fixed point iterative technique in a general inner product space. In two previous papers [6,9] the convergence properties of the minimal polynomial and reduced rank extrapolation methods, as they are applied to the vector sequence above, were analyzed. In particular, asymptotically optimal convergence results pertaining to some of the rows of the tables associated with these two methods were obtained. In the present work we continue this analysis and provide analogous results for the remaining (intermediate) rows of these tables. In particular, when {x _m} _m ^=0/ is a convergent sequence, the main result of this paper says, roughly speaking, that all of the rows converge, and it also gives the rate of convergence for each row. The results are demonstrated numerically through an example.     

Extended large deviations

LetX ₁,X ₂,..., be a sequence ofi.i.d. random variables with a moment generating function finite in a neighborhood of 0. Further, for each integern1, letS _n denote the sum of the firstn terms in this sequence. We study the extended large deviation of such sums, meaning,P{S _n >n _n }, where _n is any sequence converging to infinity. We also derive functional extended large deviation theorems and then apply them to obtain functional versions of the Erdös-Rényi strong law of large numbers.Research partially supported by an NSF Grant.     

Limit sets and strengths of convergence for sequences in the duals of thread-like Lie groups

We consider a properly converging sequence of non-characters in the dual space of a thread-like group and investigate the limit set and the strength with which the sequence converges to each of its limits. We show that, if (π _k ) is a properly convergent sequence of non-characters in , then there is a trade-off between the number of limits σ which are not characters, their degrees, and the strength of convergence i _σ to each of these limits (Theorem 3.2). This enables us to describe various possibilities for maximal limit sets consisting entirely of non-characters (Theorem 4.6). In Sect. 5, we show that if (π _k ) is a properly converging sequence of non-characters in and if the limit set contains a character then the intersection of the set of characters (which is homeomorphic to ) with the limit set has at most two components. In the case of two components, each is a half-plane. In Theorem 7.7, we show that if a sequence has a character as a cluster point then, by passing to a properly convergent subsequence and then a further subsequence, it is possible to find a real null sequence (c _k ) (with ) such that, for a in the Pedersen ideal of C ^*(G _N ), exists (not identically zero) and is given by a sum of integrals over .     

A Class of Strongly Decomposable Abelian Groups

Let G be a completely decomposable torsion-free Abelian group and G= G_i, where G _i is a rank 1 group. If there exists a strongly constructive numbering of G such that (G,) has a recursively enumerable sequence of elements g _i G _i , then G is called a strongly decomposable group. Let pi, i, be some sequence of primes whose denominators are degrees of a number p _i and let . A characteristic of the group A is the set of all pairs ‹ p,k› of numbers such that for some numbers i ₁,...,i _k . We bring in the concept of a quasihyperhyperimmune set, and specify a necessary and sufficient condition on the characteristic of A subject to which the group in question is strongly decomposable. Also, it is proved that every hyperhyperimmune set is quasihyperhyperimmune, the converse being not true.     

Limit distribution of the last exit time for stationary random sequences

Summary For a strictly stationary random sequence (X _i) _i0 we find sufficient conditions such that the distribution of the last exit time t = max{i X _i>i} (>0) tends weakly to a nondegenerate limit distribution as 0.     

Parametrized Partitions of Products of Finite Sets*

For every infinite sequence of positive integers and every Borel partition c : ×[]{0, 1} there is H[] and a sequence of subsets of , with |H_i|=m_i for every i, such that c is constant on .* Research partially supported by CNRS-FONACIT Project PI 2000001471. This author thanks the University of Paris VII for hospitality.     

The Singularity Spectrum f(α) of Some Moran Fractals

We show that the multifractal decomposition behaves as expected for a family of sets E known as homogeneous Moran fractals associated with the Fibonacci sequence , using probability measures () associated with the Fibonacci sequence . For each value of a parameter (_min, _max), we define multifractal components E of E, and show that they are fractals in the sense of Taylor. We give the explicit formula for the dimension of E. Also our method can be used for the Moran fractals associated with some more general sequences.     

On Lower Tail Probabilities of Positive Random Sums

Let be a sequence of independent identically distributed positive random variables with O-regularly varying distribution F at 0. Given a sequence of positive numbers, we show that belongs to the Type I domain of attraction of extremes for minima, by means of relating the asymptotic behaviour of P{S < } as 0, to that of E{e^-S/}. Our contribution is that we dispense with the unnatural moment condition from the literature, that F has finite variance. This in turn permits a novel application to lower tails of -stable distributions on Hilbert space.AMS 2000 Subject Classification. Primary—60G50, 60G70, Secondary—60B12, 60E07, 60F05, 60G52Research supported by NFR Grant M 650-19981841/2000, and by M.R. Leadbetter     

Sur l'équivalence de certaines mesures produit

Sommaire SoitG={g _k ,kN} une suite de variables aléatoires gaussiennes centrées réduites et indépendantes; soit de plusY={y _k ,kN} une suite indépendante deG de variables aléatoires indépendantes. On étudie à quelles conditions la loi deG+Y est équivalente à celle deG. On utilise pour cela les lois zéro-un vérifiées parG en analysant leurs effets, maximaux sur la loi deY.

---

Summary LetG={g _k ,kN} be a sequence of independent Gaussian centred reduced random variables; let moreoverY={y _k ,kN} be a sequence independent ofG of independent random variables: For obtaining conditions characterizing the equivalence of the distributions ofG andG+Y, we use the zero-one laws verified byG, first for the convergence of the series _k g _k or _k (g _k ² –a _k ), secundly for the asymptotic behavior of the sequence {g _k ,kN} and we analyze their maximal effects on the distribution ofY.

     

An Extension of Sums of Squares Relaxations to Polynomial Optimization Problems Over Symmetric Cones

This paper is based on a recent work by Kojima which extended sums of squares relaxations of polynomial optimization problems to polynomial semidefinite programs. Let and be a finite dimensional real vector space and a symmetric cone embedded in ; examples of and include a pair of the N-dimensional Euclidean space and its nonnegative orthant, a pair of the N-dimensional Euclidean space and N-dimensional second-order cones, and a pair of the space of m × m real symmetric (or complex Hermitian) matrices and the cone of their positive semidefinite matrices. Sums of squares relaxations are further extended to a polynomial optimization problem over , i.e., a minimization of a real valued polynomial a(x) in the n-dimensional real variable vector x over a compact feasible region , where b(x) denotes an - valued polynomial in x. It is shown under a certain moderate assumption on the -valued polynomial b(x) that optimal values of a sequence of sums of squares relaxations of the problem, which are converted into a sequence of semidefinite programs when they are numerically solved, converge to the optimal value of the problem. Research supported by Grant-in-Aid for Scientific Research on Priority Areas 16016234.     

Monodromy Approach to the Scaling Limits in Isomonodromy Systems

The isomonodromy deformation method is applied to the scaling limits in the linear N×N matrix equations with rational coefficients to obtain the deformation equations for the algebraic curves that describe the local behavior of the reduced versions for the relevant isomonodromy deformation equations. The approach is illustrated by the study of the algebraic curve associated with the n-large asymptotics in the sequence of the biorthogonal polynomials with cubic potentials.     

On sequences of Dirichlet polynomials uniformly bounded on half planes

Given and a sequence of Dirichlet polynomials estimates for the coefficientsa _n are proved if {_n} is uniformly bounded on a region containing a half plane. Thereby a result is obtained which is an analogue of a known result for polynomials, that is for theA-transforms of the geometric sequence; moreover a Jentzsch type theorem for {_n(z)} is derived.     

Limit of Solutions of a SDE with a Large Drift Driven by a Poisson Random Measure

We consider a sequence of {X _n} of R ^d-valued processes satisfying a stochastic differential equation driven by a Brownian motion and a compensated Poisson random measure, with _n ~ _n with a large drift. Let be a m-dimensional submanifold (m<d), where F vanishes. Then under some suitable growth conditions for _n ~ _n, and some conditions for F, we show that dist(X _n, )0 before it exits any given compact set, that is, the large drift term forces X _n close to . And if the coefficients converge to some continuous functions, any limit process must actually stay on and satisfy a certain stochastic differential equation driven by Brownian motion and white noise.