强相依非平稳序列上超点过程的收敛性

{ξi,i≥1}为标准化的正态序列,rij=Cov(ξi,ξj)。M(nk)是{ξi,i≥1}第k个最大值,本文在条件:j-i→∞时r■log(j-i)→γ∈(0,∞)下,得到了ξ1,ξ2,…,ξn时间正规化上超水平u(n1),u(n2),…,u(n)r形成的点过程依分布收敛到定义在(0,+∞)×R上的二维Cox-过程。     

关于随机过程依分布收敛的一点注记

设(Ω,T,P)为一个概率空间,ξ_n和ξ是概率空间上的随机过程。经常遇到的问题是:已知随机过程序列{ξ_n}的有限维分布收敛于ξ的有限维分布,问还要加些什么条件,方能保证{ξ_n}依分布收敛于ξ?这类问题可化为函数空间上测度序列弱收敛问题。[3]给出距离空间(x,ρ)上测度弱收敛的一个充分条件如下。[定理]设(x,ρ)为距离空间,B_x=σ(全体开集),T为单位区间I的不空子集。     

平稳高斯过程最大值的极限分布

设{(ξt),t≥0}为平稳高斯过程,E((ξt))=0,E(2ξ(t))=1,E(ξ(0)(ξt))=r(t).当r(t)logt r∈(0,∞),且r(t)单调下降到零时,得到了M(T)=sup{ξ(t);0≤t≤T}的极限分布.     

NSD随机阵列加权和最大值的收敛性

利用负超可加相依(NSD)随机阵列的Rosenthal型矩不等式和截尾方法, 在随机阵列{X_nk, 1≤k≤k_n, n≥1}关于{a_nk, 1≤k≤k_n, n≥1}一致可积的条件下, 讨论NSD随机阵列加权和最大值的弱收敛、 L^r收敛和完全收敛性.     

强一致收敛下弱几乎周期点和周期序列跟踪性的研究

在强一致收敛下,研究了弱几乎周期点和周期序列跟踪性,得到弱几乎周期点和周期序列跟踪性的若干结论: (1)设序列映射{f_n}强一致收敛于等度连续映射f,且点列{x_k}是每个映射f_n的弱几乎周期点. 若$\mathop {\lim }\limits_{k \to \infty } {x_k} = x$,则x是f的弱几乎周期点. (2)若序列映射{f_n}强一致收敛于等度连续映射f,则limsup W(f_n)?W(f). (3)若f_n具有fine周期序列跟踪性,则f具有周期序列跟踪性.     

四阶四点奇异边值问题的可解性

利用Leray-Sachuder原理研究了一类四阶四点边值问题:{u(4)(t)+f(t,u(t),u″(t)=0,t∈(0,1);u(0) = 0,u(1) = au(η),u″(0) = 0,u(1) = bu(ξ).其中,η,ξ∈[0,1],a,b≥0且满足0≤aη≤1,0≤bξ≤1,得到其解的存在性,放宽利用上下解时对函数f(t,u,v)单调性的限制,并且在t=0或t=1有奇异性.     

由强混合序列生成的线性过程重对数律的精确渐近性质

设{ε_t,t∈Z}为定义在同一概率空间(Ω,F,P )上的严平稳随机变量序列, 满足Eε₀=0, E|ε₀|^p<∞, 对某个p>2, 且满足强混合条件. {a_j, j∈Z}为一实数序列, 利用由强混合序列生成的线性过程的弱收敛定理及矩不等式讨论了在b_n=O(1/log log n)的条件下的一类加权级数的收敛性质.     

三角代数上互逆元处的高阶ξ-Lie可导映射

设U=Tri(A,M,B )是含单位元1的三角代数,1_A、1_B分别是A和B的单位元。对任意的A∈A, B∈B分别存在整数k₁、k₂,使得k₁1_A-A, k₂1_B-B在三角代数中可逆。利用代数分解的方法,证明了如果{φ_n}_n∈N:U→U是一列线性映射满足对任意的U,V∈U且UV=VU=1,有φ_n(［U,V］_ξ)=∑_i+j=nφ_i(U)φ_j(V)-ξφ_i(V)φ_j(U)(ξ≠0,1),则{φ_n}_n∈N是U上的高阶导子,其中φ₀=id₀是恒等映射,［U,V］_ξ=UV-ξVU。     

三阶非线性微分方程边值问题正解的存在性

本文应用锥上的不动点定理研究了三阶四点边值问题{u'(t)+f(t,u(t))=0,t∈[0,1],u′(0)=αu(ξ),u′(1)+βu(η)=0,u″(0)=0正解的存在性,其中α和β是正的参数,0≤ξ≤η≤1.在f满足适当的增长条件下,本文通过对核函数的上下界估计获得了该问题正解的存在性.     

可换Lipschitz半群的拟轨道的渐近动态

设C是实Hilbert空间H的闭凸集,G是含有单位元的可换半群。如果■={S(t):t∈G}是C上的Lipschitz映射所成半群,用k_t表示S(t)的Lipschitz常数,有lim supk_1≤1,那么■的拟轨道{u(t):t∈G}弱收敛于C中某一点的充分必要条件是对任意,h∈G,u(t+h)-u(t)弱收敛于0。     

由混合序列生成的线性过程加权和的极限定理

假设线性过程Xt=∑〖DD(〗∞〖〗j=0〖DD)〗ajξt-j, t≥1, 其中{ξt,t∈Z}为一零均值的混合序列, {aj, j≥0}为一实数序列, 满足∑〖DD(〗∞〖〗j=0〖DD)〗j〖JB(|〗aj〖JB)|〗<∞, {ani,1≤i≤n,n≥1}为一实值的三角阵列, 在适当的假设条件下, 利用混合序列的中心极限定理及相应的概率不等式, 证明了由混合序列生成线性过程加权和的极限定理.     

A class of quasisymmetric mappings with substantial points

It is proved that K_q(h)=K₀(h) for every h in some class of quasisymmetric mappings of the unit circle with substantial points, where K_q(h)=sup{M(h(Q))/M(Q); Q is a quadrilateral with the domain unit disk} and K₀(h) is the extremal maximum dilatation of h.     

Strong stability of linear forms in φ -mixing random variables

In this paper some new results of strong stability of linear forms in φ-mixing random variables are given. It is mainly proved that for a sequence of φ-mixing random variables {x _n , n ⩾ 1} and two sequences of positive numbers {a _n , n ⩾ 1} and {b _n , n ⩾ 1} there exist d _n ∈ R, n = 1,2,L, such that a.s. under some suitable conditions. The results extend and improve the corresponding theorems for independent identically distributed random variables. Foundation item: Supported by the National Natural Science Foundation of China(10671149)     

Generalized Galerkin approximations for pseudoinverses and operator equations of the first kind

0 IntroductionLeqtuiliHnebaer afo rHimlbaenrtd sap baoceun.dLeedtlianaenardfufnbceti oan baolu onndeHd ,serse--spectively. Our goal is to seekv∈Hsuchthata(u,v) =f(u) , u∈ H (1)As a rule, we call the expression (1) a variational equationandva solutiontothis equation.By virtue of the Riesz repre-sentationtheorem,there exists a uniquely determined boundedlinear operatorA:H→Hsuchthata(u,v) =〈u,Av〉, u,v∈ H,where〈·,·〉is the inner product onH.If the sesquilinearformais also with posit…     

Almost sure convergence of weighted sums of NA sequences

For double arrays of constants {a _ni, 1≤i≤k _n, n≥1} and NA r.v. 's {X _n, n≥1}, conditions for almost sure convergence of are given. Both casesk _n ↑ ∞ andk _n=∞ are treated. A Marcinkiewicz-type theorem for i. d. NA sequences is obtained as a special case. Supported by the National Natural Science Foundation of China Cheng Riyan: born in 1968, MS student     

Initial-boundary value problem for a class of quasilinear degenerate parabolic equations

In this work, we study the quasilinear initial-boundary value problem $\sum\limits_{i, j = 1}^m {X_i^ * A_{i j} (x, t, u)X _j u} + \sum\limits_{j = 1}^m {B_j (x, t, u)X _j u + C} (x, t, u) = 0$ , in ]0,T[×Ω;u| _t=0 =φ, onΩ;u=ψ, on [0,T[×?Ω, whereX={X ₁, ...,X _m } is a system of real smooth vector fields which is defined on an open domainM of R ⁿ , and satisfies the Hörmander's condition, Ω?M. Assume that ?Ω is non characteristic for the systemX ₁,...,X _m . Under some hypothesis for the boundary of domain and the elliptic structure condition for nonlinear coefficientsA _ij ,B _j ,C, (i, j=1, ...,m), we have proved that the existence and regularity of solution for aboveinitial-boudary value problems.     

LPQD列生成线性过程部分和的精确渐近性

设{ε_t;t∈Z^{+}是一严平稳零均值的LPQD随机变量序列, 并且02₁<∞, σ2, 0<σ2<∞, {a_j; j∈N }是一实数序列, 定义线性过程X_t. 利用弱收敛定理和矩不等式, 对一般的拟权函数和边界函数, 证明了{M_n}和{S_n}的精确渐近性.}     

带有强阻尼时滞项的m-Laplacian型波方程解的爆破

研究带有强阻尼时滞项的m-Laplacian型波方程:utt-Δmu-Δu+g*Δu-μ1Δut(x,t)-μ2Δut(x,t-τ)=(u)p-2u解的爆破:当初始能量00,ν>0,t≥0),在(0,t)...     

一类算子稳定过程图集的Hausdorff测度

设X={X（t）,t∈R＋}是取值R^d上的,指数为对角矩阵的算子稳定过程．文章将探讨X的图集的Hausdorff测度问题．更确切地,利用逗留时得到了图集G（[0,1]）={（t,X（t））：0≤t≤1}的确切Hausdorff测度函数．这一结果推广了已有文献中类似的结果．