部分函数型线性空间自回归模型的假设检验

本文提出了部分函数型线性空间自回归模型的空间效应以及参数效应的假设检验问题.首先利用函数型主成分分析方法估计斜率函数,利用广义矩估计方法估计参数.然后利用得到的相合估计,在原假设和备择假设下,构造了基于残差平方和的检验统计量,同时给出了此检验统计量的渐近性质.模拟结果表明在有限样本下,检验统计量具有良好表现.最后将部分函数型线性空间自回归模型的检验应用到一个关于经济增长的数据案例中,说明所提出的检验统计量的应用表现.     

半参数空间分位回归模型的估计与变量选择

对于高维空间数据,利用半参数空间自回归进行建模,模型中会同时存在内生性、非线性、变量过多等问题。本文研究半参数空间分位回归模型,提出了新的估计程序:首先利用样条基函数,对模型中未知平滑函数进行逼近,解决非线性问题;然后运用特征向量空间滤波,将空间滞后因子转化为空间代理变量的线性组合,有效解决了内生性问题;利用再中心化影响函数,进行无条件分位回归建模,能够刻画不同分位水平下变量之间的关系;最后引入自适应Lasso惩罚,对高维线性部分进行变量选择,得到系数的稀疏估计,有效增强了模型的可解释性。数值模拟中对参数作不同的设置,展现了本文提出方法的有效性。最后,利用半参数空间分位回归模型分析了住房销售价格数据集。     

单调函数型数据半参数模型的估计（英文）

本文研究了单调约束条件下函数型数据半参数模型的估计问题,利用非线性混合效应模型的方法,给出模型中未知参数向量和单调参数曲线的估计,并通过模拟对提出的方法进行数值分析.     

纵向数据下半参数工具变量模型的经验似然推断

本文研究纵向数据下半参数工具变量模型中回归系数的区间估计问题.首先利用B-样条方法逼近半参数模型中的非参数函数.为了处理内生变量和纵向数据的组内相关性,对模型中回归系数提出了基于工具变量和二次推断函数的有效经验对数似然比统计量,并证明所提出统计量渐近服从标准卡方分布,由此构造回归系数的置信域.     

基于联合估计方法的广义随机系数自回归模型的统计推断

本文利用联合估计函数方法(CEF)对广义随机系数自回归(GRCA)模型进行统计研究.应用联合估计函数方法得到广义随机系数自回归模型参数估计量,证明了提出的参数估计量的相合性和渐近正态性,利用数值模拟对提出的参数统计量进行对比分析,数值模拟结果表明,联合估计方法的参数估计量优于基于估计函数方法、伪极大似然方法、最小二乘方法的参数估计量,实证研究也说明CEF方法具有较好的效果.     

在函数逼近论观点下半参数变系数非线性回归函数的估计

在函数逼近论的观点下研究了半参数变系数非线性回归函数的估计问题.采用总体之下L2多项式最佳逼近的方法与样本之下矩估计的方法,独立地分别作出非参数部分与变系数参数部分的解析函数形式的估计,最终得到回归函数的L2与强相合之联合收敛意义下的估计.     

具有测量误差的非线性模型的Bayes估计

测量中大量的函数模型都是非线性回归模型．当回归变量含有一定的测量误差时，我们得到非线性测量误差模型．本讨论了这种模型中未知参数具有正态先验分布时的参数Bayes估计方法，并对这种估计进行了影响分析，证明了删除模型与均值漂移模型中参数的Bayes估计相同，利用Cook统计量给出了删除模型下参数的Bayes估计的影响度量．     

核实数据下误差在反映变量的非线性半参数模型的降维估计

考虑响应变量带有一般测量误差的非线性半参数模型.在核实数据的帮助下,利用半参数降维技术构造未知参数和非参数函数的估计.在一定条件下证明未知参数估计的渐近正态性和非参数函数估计的最优收敛速度.通过数值模拟说明所提估计方法在有限样本下的有效性.     

非线性半参数回归模型的最大经验似然估计

对于非线性半参数回归模型的估计问题,利用经验似然方法,给出了回归系数,光滑函数以及误差方差的最大经验似然估计.在一定条件下证明了所得估计量的渐近正态性和相合性.     

函数型部分线性乘积模型的B-样条估计

本文研究了函数型部分线性乘积模型,该模型可用于响应变量为正数的函数型数据的统计建模问题,经过对数变换后模型转化为函数型部分线性模型.基于B-样条,通过极小化最小一乘相对误差(LARE)和最小乘积相对误差(LPRE),分别给出模型的LARE估计和LPRE估计,其中B-样条基的维数利用Schwarz信息准则选取.对两种估计方法分别给出斜率函数估计的相合性和参数部分估计的渐近正态性,并且证明了斜率函数的收敛率达到了非参数函数估计的最优速率.蒙特卡洛模拟用来比较所提出的方法与最小一乘(LAD)估计和最小二乘(LS)估计在不同误差分布下的有限样本性质,模拟结果表明所提方法是有效和实用的.最后通过一个实际数据分析的例子来说明模型的应用.     

Local topological equivalence of nonsmooth functions in Hilbert spaces

In this paper we investigate the relation between nonsmooth functions with domain in a Hilbert space and their local approximations. We consider Lipschitz functions and define an approximation model with directional derivatives. The qualitative behaviour of the approximation is studied by means of the concept of topological equivalence. Using this concept we establish the existence of a local coordinate transformation between the original function and the local approximation.     

Convergence of non-cyclic infinite products of operators

Using a geometric approach (in particular, angles between subspaces), we establish useful estimates regarding convergence of infinite products involving orthogonal projections and other (possibly nonlinear) nonexpansive operators in Hilbert space.     

无穷维Hilbert空间中的伪单调非线性互补问题

建立伪单调非线性互补问题在实Hilbert空间任意闭凸锥上的解的存在性理论,特别把互补问题在有限维实Hilbert空间上的一些重要理论推广到无限维实Hilbert空间,还证明了解的存在的可行性理论.     

A characteristic operator function for the class of n-hypercontractions

We consider a class of bounded linear operators on Hilbert space called n-hypercontractions which relates naturally to adjoint shift operators on certain vector-valued standard weighted Bergman spaces on the unit disc. In the context of n-hypercontractions in the class C0⋅ we introduce a counterpart to the so-called characteristic operator function for a contraction operator. This generalized characteristic operator function Wn,T is an operator-valued analytic function in the unit disc whose values are operators between two Hilbert spaces of defect type. Using an operator-valued function of the form Wn,T, we parametrize the wandering subspace for a general shift invariant subspace of the corresponding vector-valued standard weighted Bergman space. The operator-valued analytic function Wn,T is shown to act as a contractive multiplier from the Hardy space into the associated standard weighted Bergman space.     

Error bounds for finite element solutions of mildly nonlinear elliptic boundary value problems

Summary The equivalence in a Hilbert space of variational and weak formulations of linear elliptic boundary value problems is well known. This same equivalence is proved here for mildly nonlinear problems where the right hand side of the differential equation involves the solution function. A finite element approximation to the solution of the weak problem ina finite dimensional subspace of the original Hilbert space is defined. An inequality bounding the error in this approximation over all functions of the space is derived, and in particular this holds for an interpolant to the weak solution. Thus this inequality, together with previously known, interpolation error bounds, produces a bound on the finite element solution to this nonlinear problem. An example of a mildly nonlinear Poisson problem is given.     

A quantitative notion of redundancy for finite frames

The objective of this paper is to improve the customary definition of redundancy by providing quantitative measures in its place, which we coin upper and lower redundancies, that match better with an intuitive understanding of redundancy for finite frames in a Hilbert space. This motivates a carefully chosen list of desired properties for upper and lower redundancies. The means to achieve these properties is to consider the maximum and minimum of a redundancy function, which is interesting in itself. The redundancy function is defined on the sphere of the Hilbert space and measures the concentration of frame vectors around each point. A complete characterization of functions on the sphere which coincide with a redundancy function for some frame is given. The upper and lower redundancies obtained from this function are then shown to satisfy all of the intuitively desirable properties. In addition, the range of values they assume is characterized.     

Parametric lyapunov function method for solving nonlinear systems in hilbert spaces

This paper considers a nonlinear systems of finite number of equations in Hilbert spaces. Using the left-hand sides of the equations, a family of scalar and convex functions is constructed. Under the hypothesis of a convexity, these functions are the Lyapunov functions for the differential inclusion for which the right-hand side is the negative value of a maximial monotone operator. Every solution of the differntial inclusion solves the nonlinear system of equations in a finite moment of the time. An iterative method for finding at least one of the solutions of the system is presented. This method is based on the Lyapunov function and it depends on a vector parameter. The suitable choice of the paremeters implies the linear or superlinear (supergeometric) convergence of the interative method     

g-拟凸函数的一个判别准则

证明了如下结果:设g∶H→H,C H是非空开的g-凸集,g(C)是凸集,f是C上的上半连续函数且存在α∈(0,1),使得f(αg(x)+(1-α)g(y))m ax{f。g(x),f。g(y)},x,y∈C,则f为C上的g-拟凸函数.     

向量值Nevanlinna理论初探

本文主要讨论定义在复平面取值为无穷维希尔伯特空间上的向量值亚纯函数的Ｎｅｖａｎｌｉｎｎａ理论．给出了推广的Ｐｏｉｓｓｏｎ－Ｊｅｎｓｅｎ－Ｎｅｖａｎｌｉｎｎａ公式以及推广的第一基本定理的证明，并对向量值特征函数的性质予以讨论     

Evolution equations associated with non-isothermal phase separation: Subdifferential approach

This paper is concerned with a non-isothermal phase separation model, for instance, describing component separation dynamics for a class of binary systems (solid-solid systems). Our model is a couple of nonlinear parabolicPDE's which are possibly degenerate and singular at the same time. It will be shown that our model can be reformulated as a nonlinear evolution equation governed by subdifferential operator in a Hilbert space, and its mathematically basic questions, such as existence and uniqueness, can be handled in the abstract theory of nonlinear evolution equations in Hilbert spaces. Entrata in Redazione il 15 luglio 1997 e, in versione riveduta, il 25 ottobre 1997.