共查询到20条相似文献,搜索用时 484 毫秒
1.
Let Ωm be the set of partitions, ω, of a finite m-element set; induce a uniform probability distribution on Ωm, and define Xms(ω) as the number of s-element subsets in ω. We alow the existence of an integer-valued function n=n(m)(t), t?[0, 1], and centering constants bms, 0?s? m, such that converges to the ‘Brownian Bridge’ process in terms of its finite-dimensional distributions. 相似文献
2.
Malcolm R. Adams 《Journal of Functional Analysis》1983,52(3):420-441
Let Q be a self-adjoint, classical, zeroth order pseudodifferential operator on a compact manifold X with a fixed smooth measure dx. We use microlocal techniques to study the spectrum and spectral family, {ES}S∈ as a bounded operator on L2(X, dx).Using theorems of Weyl (Rend. Circ. Mat. Palermo, 27 (1909), 373–392) and Kato (“Perturbation Theory for Linear Operators,” Springer-Verlag, 1976) on spectra of perturbed operators we observe that the essential spectrum and the absolutely continuous spectrum of Q are determined by a finite number of terms in the symbol expansion. In particular SpecESSQ = range(q(x, ξ)) where q is the principal symbol of Q. Turning the attention to the spectral family {ES}S∈, it is shown that if is considered as a distribution on ×X×X it is in fact a Lagrangian distribution near the set where (s, x, y, σ, ξ,η) are coordinates on T1(×X×X) induced by the coordinates (s, x, y) on ×X×X. This leads to an easy proof that is a pseudodifferential operator if ?∈C∞() and to some results on the microlocal character of Es. Finally, a look at the wavefront set of leads to a conjecture about the existence of absolutely continuous spectrum in terms of a condition on q(x, ξ). 相似文献
3.
Let (Xn)n? be a sequence of real, independent, not necessarily identically distributed random variables (r.v.) with distribution functions FXn, and Sn = Σi=1nXi. The authors present limit theorems together with convergence rates for the normalized sums ?(n)Sn, where ?: → +, ?(n) → 0, n → ∞, towards appropriate limiting r.v. X, the convergence being taken in the weak (star) sense. Thus higher order estimates are given for the expression ∝f(x) d[F?(n)Sn(x) ? FX(x)] which depend upon the normalizing function ?, decomposability properties of X and smoothness properties of the function f under consideration. The general theorems of this unified approach subsume O- and o-higher order error estimates based upon assumptions on associated moments. These results are also extended to multi-dimensional random vectors. 相似文献
4.
Robert Chen 《Journal of multivariate analysis》1978,8(2):328-333
Let {Xn}n≥1 be a sequence of independent and identically distributed random variables. For each integer n ≥ 1 and positive constants r, t, and ?, let Sn = Σj=1nXj and . In this paper, we prove that (1) lim?→0+?α(r?1)E{N∞(r, t, ?)} = K(r, t) if E(X1) = 0, Var(X1) = 1, and E(| X1 |t) < ∞, where 2 ≤ t < 2r ≤ 2t, , and ; (2) if 2 < t < 4, E(X1) = 0, Var(X1) > 0, and E(|X1|t) < ∞, where G(t, ?) = E{N∞(t, t, ?)} = Σn=1∞nt?2P{| Sn | > ?n} → ∞ as ? → 0+ and , i.e., H(t, ?) goes to infinity much faster than G(t, ?) as ? → 0+ if 2 < t < 4, E(X1) = 0, Var(X1) > 0, and E(| X1 |t) < ∞. Our results provide us with a much better and deeper understanding of the tail probability of a distribution. 相似文献
5.
Loren D. Pitt 《Journal of multivariate analysis》1978,8(1):45-54
For Gaussian vector fields {X(t) ∈ Rn:t ∈ Rd} we describe the covariance functions of all scaling limits Y(t) = limα↓0 B?1(α) X(αt) which can occur when B(α) is a d × d matrix function with B(α) → 0. These matrix covariance functions are found to be homogeneous in the sense that for some matrix L and each α > 0, . Processes with stationary increments satisfying (1) are further analysed and are found to be natural generalizations of Lévy's multiparameter Brownian motion. 相似文献
6.
For fixed p (0 ≤ p ≤ 1), let {L0, R0} = {0, 1} and X1 be a uniform random variable over {L0, R0}. With probability p let {L1, R1} = {L0, X1} or = {X1, R0} according as ; with probability 1 ? p let {L1, R1} = {X1, R0} or = {L0, X1} according as , and let X2 be a uniform random variable over {L1, R1}. For n ≥ 2, with probability p let {Ln, Rn} = {Ln ? 1, Xn} or = {Xn, Rn ? 1} according as , with probability 1 ? p let {Ln, Rn} = {Xn, Rn ? 1} or = {Ln ? 1, Xn} according as , and let Xn + 1 be a uniform random variable over {Ln, Rn}. By this iterated procedure, a random sequence {Xn}n ≥ 1 is constructed, and it is easy to see that Xn converges to a random variable Yp (say) almost surely as n → ∞. Then what is the distribution of Yp? It is shown that the Beta, (2, 2) distribution is the distribution of Y1; that is, the probability density function of Y1 is g(y) = 6y(1 ? y) I0,1(y). It is also shown that the distribution of Y0 is not a known distribution but has some interesting properties (convexity and differentiability). 相似文献
7.
Let {Xn} be a stationary Gaussian sequence with E{X0} = 0, {X20} = 1 and E{X0Xn} = rnn Let cn = (2ln n), bn = cn? c-1n ln(4π ln n), and set Mn = max0 ?k?nXk. A classical result for independent normal random variables is that Berman has shown that (1) applies as well to dependent sequences provided rnlnn = o(1). Suppose now that {rn} is a convex correlation sequence satisfying rn = o(1), (rnlnn)-1 is monotone for large n and o(1). Then for all x, where Ф is the normal distribution function. While the normal can thus be viewed as a second natural limit distribution for {Mn}, there are others. In particular, the limit distribution is given below when rn is (sufficiently close to) γ/ln n. We further exhibit a collection of limit distributions which can arise when rn decays to zero in a nonsmooth manner. Continuous parameter Gaussian processes are also considered. A modified version of (1) has been given by Pickands for some continuous processes which possess sufficient asymptotic independence properties. Under a weaker form of asymptotic independence, we obtain a version of (2). 相似文献
8.
B.G. Pittel 《Stochastic Processes and their Applications》1980,10(1):33-48
Let X1,X2,… be i.i.d. random variables with a continuous distribution function. Let R0=0, Rk=min{j>Rk?1, such that Xj>Xj+1}, k?1. We prove that all finite-dimensional distributions of a process , converge to those of the standard Brownian motion. 相似文献
9.
The probability measure of X = (x0,…, xr), where x0,…, xr are independent isotropic random points in n (1 ≤ r ≤ n ? 1) with absolutely continuous distributions is, for a certain class of distributions of X, expressed as a product measure involving as factors the joint probability measure of (ω, ?), the probability measure of p, and the probability measure of . Here ω is the r-subspace parallel to the r-flat η determined by X, ? is a unit vector in ω⊥ with ‘initial’ point at the origin [ω⊥ is the (n ? r)-subspace orthocomplementary to ω], p is the norm of the vector z from the origin to the orthogonal projection of the origin on η, and , where α is a scale factor determined by p. The probability measure for ω is the unique probability measure on the Grassmann manifold of r-subspaces in n invariant under the group of rotations in n, while the conditional probability measure of ? given ω is uniform on the boundary of the unit (n ? r)-ball in ω⊥ with centre at the origin. The decomposition allows the evaluation of the moments, for a suitable class of distributions of X, of the r-volume of the simplicial convex hull of {x0,…, xr} for 1 ≤ r ≤ n. 相似文献
10.
P. Révész 《Stochastic Processes and their Applications》1983,15(2):169-179
Let U1, U2,… be a sequence of independent, uniform (0, 1) r.v.'s and let R1, R2,… be the lengths of increasing runs of {Ui}, i.e., X1=R1=inf{i:Ui+1<Ui},…, Xn=R1+R2+?+Rn=inf{i:i>Xn?1,Ui+1<Ui}. The first theorem states that the sequence can be approximated by a Wiener process in strong sense.Let τ(n) be the largest integer for which R1+R2+?+Rτ(n)?n, and . Here Mn is the length of the longest increasing block. A strong theorem is given to characterize the limit behaviour of Mn.The limit distribution of the lengths of increasing runs is our third problem. 相似文献
11.
Let be the Schwartz space of rapidly decreasing real functions. The dual space 1 consists of the tempered distributions and the relation ? L2() ? 1 holds. Let γ be the Gaussian white noise on 1 with the characteristic functional , ξ ∈ , where ∥·∥ is the L2()-norm. Let ν be the Poisson white noise on 1 with the characteristic functional = exp?∫ {[exp(iξ(t)u)] ? 1 ? (1 + u2)?1(iξ(t)u)} dη(u)dt), ξ ∈ , where the Lévy measure is assumed to satisfy the condition ∫u2dη(u) < ∞. It is proved that γ1ν has the same dichotomy property for shifts as the Gaussian white noise, i.e., for any ω ∈ 1, the shift of γ1ν by ω and γ1ν are either equivalent or orthogonal. They are equivalent if and only if when ω ∈ L2() and the Radon-Nikodym derivative is derived. It is also proved that for the Poisson white noice νω is orthogonal to ν for any non-zero ω in 1. 相似文献
12.
We derive sufficient conditions for ∝ λ (dx)6Pn(x, ·) - π6 to be of order o(ψ(n)-1), where Pn (x, A) are the transition probabilities of an aperiodic Harris recurrent Markov chain, π is the invariant probability measure, λ an initial distribution and ψ belongs to a suitable class of non-decreasing sequences. The basic condition involved is the ergodicity of order ψ, which in a countable state space is equivalent to for some i, where τi is the hitting time of the tate i. We also show that for a general Markov chain to be ergodic of order ψ it suffices that a corresponding condition is satisfied by a small set.We apply these results to non-singular renewal measures on providing a probabilisite method to estimate the right tail of the renewal measure when the increment distribution F satisfies ∝ tF(dt) 0; > 0 and ∝ ψ(t)(1- F(t))dt< ∞. 相似文献
13.
Let (μt)∞t=0 be a k-variate (k?1) normal random walk process with successive increments being independently distributed as normal N(δ, R), and μ0 being distributed as normal N(0, V0). Let Xt have normal distribution N(μt, Σ) when μt is given, t = 1, 2,….Then the conditional distribution of μt given X1, X2,…, Xt is shown to be normal N(Ut, Vt) where Ut's and Vt's satisfy some recursive relations. It is found that there exists a positive definite matrix V and a constant θ, 0 < θ < 1, such that, for all t?1, where the norm |·| means that |A| is the largest eigenvalue of a positive definite matrix A. Thus, Vt approaches to V as t approaches to infinity. Under the quadratic loss, the Bayesian estimate of μt is Ut and the process {Ut}∞t=0, U0=0, is proved to have independent successive increments with normal N(θ, Vt?Vt+1+R) distribution. In particular, when V0 =V then Vt = V for all t and {Ut}∞t=0 is the same as {μt}∞t=0 except that U0 = 0 and μ0 is random. 相似文献
14.
E.B Dynkin 《Journal of Functional Analysis》1985,62(3):397-434
Let Xt be the Brownian motion in d. The random set Γ = {(t1,…, tn, z): Xtl = ··· = Xtn = z} in d + n is empty a.s. except in the following cases: (a) n = 1, d = 1, 2,…; (b) d = 2, n = 2, 3,…; (c) d = 3, n = 2. In each of these cases, a family of random measures Mλ concentrated on Γ is constructed (λ takes values in a certain class of measures on d). Measures Mλ characterize the time-space location of self-intersections for Brownian paths. If n = d = 1, then Mλ(dt, dz) = λ(dz) Nz(dt) where N2 is the local time at z. In the case n = 2, the set Γ can be identified with the set of Brownian loops. The measure Mλ “explodes” on the diagonal {t1 = t2} and, to study small loops, a random distribution which regularizes Mλ is constructed. 相似文献
15.
Richard L Tweedie 《Journal of Mathematical Analysis and Applications》1977,60(1):280-291
Suppose {Pn(x, A)} denotes the transition law of a general state space Markov chain {Xn}. We find conditions under which weak convergence of {Xn} to a random variable X with law L (essentially defined by ∝ Pn(x, dy) g(y) → ∝ L(dy) g(y) for bounded continuous g) implies that {Xn} tends to X in total variation (in the sense that , which then shows that L is an invariant measure for {Xn}. The conditions we find involve some irreducibility assumptions on {Xn} and some continuity conditions on the one-step transition law {P(x, A)}. 相似文献
16.
Let (Ω, , μ) be a measure space, a separable Banach space, and 1 the space of all bounded conjugate linear functionals on . Let f be a weak1 summable positive B(1)-valued function defined on Ω. The existence of a separable Hilbert space , a weakly measurable B()-valued function Q satisfying the relation is proved. This result is used to define the Hilbert space L2,f of square integrable operator-valued functions with respect to f. It is shown that for B+(1)-valued measures, the concepts of weak1, weak, and strong countable additivity are all the same. Connections with stochastic processes are explained. 相似文献
17.
Shelby J. Haberman 《Journal of multivariate analysis》1980,10(3):398-404
Let Y be an N(μ, Σ) random variable on Rm, 1 ≤ m ≤ ∞, where Σ is positive definite. Let C be a nonempty convex set in Rm with closure . Let (·,-·) be the Eculidean inner product on Rm, and let μc be the conditional expected value of Y given Y ∈ C. For v ∈ Rm and s ≥ 0, let βs(v) be the expected value of |(v, Y) ? (v, μ)|s and let γs(v) be the conditional expected value of |(v, Y) ? (v, μc)|s given Y ∈ C. For s ≥ 1, γs(v) < βs(v) if and only if , and γs(v) < βs(v) for all v ≠ 0 if and only if for any v ∈ Rm such that v ≠ 0. 相似文献
18.
Let (m?n) denote the linear space of all m × n complex or real matrices according as = or . Let c=(c1,…,cm)≠0 be such that c1???cm?0. The c-spectral norm of a matrix A?m×n is the quantity . where σ1(A)???σm(A) are the singular values of A. Let d=(d1,…,dm)≠0, where d1???dm?0. We consider the linear isometries between the normed spaces and , and prove that they are dual transformations of the linear operators which map (d) onto (c), where . 相似文献
19.
20.
Norbert Herrndorf 《Journal of multivariate analysis》1984,15(1):141-146
In this note a functional central limit theorem for ?-mixing sequences of I. A. Ibragimov (Theory Probab. Appl.20 (1975), 135–141) is generalized to nonstationary sequences (Xn)n ∈ , satisfying some assumptions on the variances and the moment condition for some b > 0, ? > 0. 相似文献