Limit theorems for sums of dependent random variables occurring in statistical mechanics

Summary We study the asymptotic behavior of partial sums S _nfor certain triangular arrays of dependent, identically distributed random variables which arise naturally in statistical mechanics. A typical result is that under appropriate assumptions there exist a real number m, a positive real number , and a positive integer k so that (S _n–nm)/n^1–1/2k converges weakly to a random variable with density proportional to exp(–¦s¦ ^2k/(2k)!). We explain the relation of these results to topics in Gaussian quadrature, to the theory of moment spaces, and to critical phenomena in physical systems.Alfred P. Sloan Research Fellow. Research supported in part by a Broadened Faculty Research Grant at the University of Massachusetts and by National Science Foundation Grant MPS 76-06644Research supported in part by National Science Foundation Grants MPS 74-04870 A01 and MCS 77-20683     

Limit theorems for sums of dependent random variables

Summary In Lai and Stout [7] the upper half of the law of the iterated logarithm (LIL) is established for sums of strongly dependent stationary Gaussian random variables. Herein, the upper half of the LIL is established for strongly dependent random variables {X _i} which are however not necessarily Gaussian. Applications are made to multiplicative random variables and to f(Z _i ) where the Z _i are strongly dependent Gaussian. A maximal inequality and a Marcinkiewicz-Zygmund type strong law are established for sums of strongly dependent random variables X _i satisfying a moment condition of the form E¦S _a,n ¦^pg(n), where , generalizing the work of Serfling [13, 14].Research supported by the National Science Foundation under grant NSF-MCS-78-09179Research supported by the National Science Foundation under grant NSF-MCS-78-04014     

Limit theorems for sums of dependent random variables

Lithuanian Mathematical Journal -     

Limit theorems for maxima of some dependent random sums

A family of extrema having the form $$Y_{mn} = \mathop {\max }\limits_{1 \leqslant i \leqslant m} \sum\limits_{j = 1}^n {X_{ij} , m,n \geqslant 1,}$$ is considered, here the random variables {X _ij }, i ?? 1, j ?? 1, are dependent in columns (with identical j) and independent in rows (with different j). The asymptotics of Y _mn for m, n ?? ?? is studied. Three particular cases are considered: a normal distribution, a Laplace distribution, and an ??-stable distribution.     

Limit theorems for sums of independent random variables defined on a nonrecurrent random walk

Let {X_i} _i= -, {_i} _i=1 be two independent sequences of randoms variables, where the _i are identically distributed and assume integer values. LetIn the paper the question of the asymptotic behavior as n of the quantity is considered. It is shown that the distribution of W_n converges to the distribution of the normal law and that the estimate of the rate of convergence has the same order as the classical estimate of Berry-Esseen.Translated from Zapiski Nauchnykh Seminarov Leningradskogo Otdeleniya Matematicheskogo Instituta im. V. A. Steklova AN SSSR, Vol. 85, pp. 17–29, 1979.The author is grateful to I. A. Ibragimov for his attention to the present work and for useful discussions.     

Limit theorems for random processes constructed from sums of independent identically distributed random variables

There are proved limit theorems for random processes constructed from sums of independent identically distributed random variables.Translated from Ukrainskii Matematicheskii Zhurnal, Vol. 43, No. 1, pp. 141–145, January, 1991.     

Limit theorems for sums of random exponentials

We study limiting distributions of exponential sums as t→∞, N→∞, where (X_i) are i.i.d. random variables. Two cases are considered: (A) ess sup X_i = 0 and (B) ess sup X_i = ∞. We assume that the function h(x)= -log P{X_i>x} (case B) or h(x) = -log P {X_i>-1/x} (case A) is regularly varying at ∞ with index 1 < ϱ <∞ (case B) or 0 < ϱ < ∞ (case A). The appropriate growth scale of N relative to t is of the form , where the rate function H₀(t) is a certain asymptotic version of the function (case B) or (case A). We have found two critical points, λ₁<λ₂, below which the Law of Large Numbers and the Central Limit Theorem, respectively, break down. For 0 < λ < λ₂, under the slightly stronger condition of normalized regular variation of h we prove that the limit laws are stable, with characteristic exponent α = α (ϱ, λ) ∈ (0,2) and skewness parameter β ≡ 1.Research supported in part by the DFG grants 436 RUS 113/534 and 436 RUS 113/722.Mathematics Subject Classification (2000): 60G50, 60F05, 60E07     

Limit theorems for maximum functionals of asymptotically normal sums of independent random variables

Translated from Problemy Ustoichivosti Stokhasticheskikh Modelei — Trudy Seminara, pp. 91–100, 1985.     

Almost sure limit theorems for random sums of multiindex random variables

In this paper, we obtain an almost sure functional limit theorem for random sums of multiindex random variables.     

One-sided limit theorems for sums of multidimensional indexed random variables

Let {X,X _n,nZ ₊ ^d } be a sequence of independent and identically distributed random variables and {a _n ,n Z ₊ ^d } be a sequence of constants. We examine the almost sure limiting behavior of weighted partial sums of the form _|n|N a _n X _n . Suppose further that eitherEX=0 orE|X|=. In most situations these normalized partial sums fail to have a limit, no matter which normalizing sequence we choose. Thus, the investigation lends itself to the study of the limit inferior and limit superior of these sequences. On the way to proving results of this type we first establish several weak laws. These weak laws prove to be of great value in establishing generalized laws of the iterated logarithm.