等式约束优化一个修正的投影变尺度法

本文研究了等式约束优化问题.利用罚函数和投影变尺度方法,得到了一个修正的算法及其全局收敛与超线性收敛率.改进了文献[J]中的方法.     

变系数测量误差模型的B-样条估计(英文)

本文考虑变系数测量误差模型的估计问题,得到该模型变系数函数修正的最小二乘B-样条估计,同时得到非参数函数估计的最优收敛速度.模拟结果表明该方法是有效的.     

Linex损失及PA样本下单边截断型分布族参数函数的EB估计

在Linex损失函数下,运用同分布PA样本密度函数的核估计方法,构造了一类单边截断型分布族参数函数的EB估计,并建立了它的收敛速度.在一定条件下,证明了这个收敛速度可充分接近1.     

带超级光滑噪声密度函数的小波自适应点态估计

利用小波方法在局部Holder空间中研究一类反卷积密度函数的点态估计问题.首先,针对超级光滑噪声给出该模型任一估计器的点态风险下界;其次,构造有限求和小波估计器,并证明其在超级光滑噪声条件下达到了最优收敛阶,即该估计器在点态风险下的收敛速度与下界一致.最后,还讨论了这类小波估计器的强收敛性.值得指出的是上述估计都是自适应的.     

自由支配集下近似平衡约束向量优化问题的稳定性研究

在自由支配集下,对一类近似平衡约束向量优化问题（AOPVF）的稳定性进行研究.首先,在较弱的凸性假设下获得了约束集映射的Berge-半连续性和约束集的闭性、凸性和紧性结果.然后,在目标函数列Gamma-收敛的假设下,分别得到了AOPVF弱有效解映射Berge 半连续和弱有效解集下Painlevé-Kuratowski收敛的充分条件,并给出例子说明结论是新颖和有意义的.     

变系数部分线性测量误差模型的B-样条估计（英文）

本文研究变系数部分线性测量误差模型的估计问题.利用纠偏方法,获得参数分量修正的最小二乘估计和非参数分量的B-样条估计.证明参数估计是相合的,渐近正态的;系数函数的B-样条估计达到非参数回归估计的最优收敛速度.模拟结果表明该方法是有效的.     

半参数面板空间滞后模型的两阶段最小二乘估计及应用

发展了一种半参数面板空间滞后模型的两阶段最小二乘估计方法.证明了参数分量估计具有渐近正态性且收敛速度为n~(-1/2),非参数分量估计在内点处具有渐近正态性,其收敛速度达到了非参数函数估计的最优收敛速度.并将方法应用于外商直接投资对劳动收入份额的影响分析.     

函数系数线性自回归模型的样条估计

基于多项式样条全局光滑方法,建立函数系数线性自回归模型中系数函数的样条估计.在适当条件下,证明了系数函数多项式样条估计的相合性,并给出了它们的收敛速度.模拟例子验证了理论结果的正确性.     

函数型部分线性复合分位数回归模型的估计（英文）

本文研究函数型部分线性复合分位数回归模型的估计问题.我们采用函数型主成分分析方法分析斜率函数,回归样条逼近非参数函数.在相当宽松的条件下给出斜率函数和非参数函数的收敛速度.最后通过理论模拟和实例分析来评价我们提出的方法.     

互补约束问题的一种新松弛规划及其收敛性

在G.H.Lin与M.Fukushima思想的启发下,针对一般形式的互补约束问题,本文构造了一种新的松弛规划.通过修正和简化G.H.Lin与M.Fukushima的证明方法,在比其更弱的假设条件下获得了该松弛规划的收敛性质.     

A class of nonlinear Lagrangians for nonconvex second order cone programming

This paper focuses on the study of a class of nonlinear Lagrangians for solving nonconvex second order cone programming problems. The nonlinear Lagrangians are generated by Löwner operators associated with convex real-valued functions. A set of conditions on the convex real-valued functions are proposed to guarantee the convergence of nonlinear Lagrangian algorithms. These conditions are satisfied by well-known nonlinear Lagrangians appeared in the literature. The convergence properties for the nonlinear Lagrange method are discussed when subproblems are assumed to be solved exactly and inexactly, respectively. The convergence theorems show that, under the second order sufficient conditions with sigma-term and the strict constraint nondegeneracy condition, the algorithm based on any of nonlinear Lagrangians in the class is locally convergent when the penalty parameter is less than a threshold and the error bound of solution is proportional to the penalty parameter. Compared to the analysis in nonlinear Lagrangian methods for nonlinear programming, we have to deal with the sigma term in the convergence analysis. Finally, we report numerical results by using modified Frisch’s function, modified Carroll’s function and the Log-Sigmoid function.     

A highly accurate collocation Trefftz method for solving the Laplace equation in the doubly connected domains

A highly accurate new solver is developed to deal with the Dirichlet problems for the 2D Laplace equation in the doubly connected domains. We introduce two circular artificial boundaries determined uniquely by the physical problem domain, and derive a Dirichlet to Dirichlet mapping on these two circles, which are exact boundary conditions described by the first kind Fredholm integral equations. As a direct result, we obtain a modified Trefftz method equipped with two characteristic length factors, ensuring that the new solver is stable because the condition number can be greatly reduced. Then, the collocation method is used to derive a linear equations system to determine the unknown coefficients. The new method possesses several advantages: mesh‐free, singularity‐free, non‐illposedness, semi‐analyticity of solution, efficiency, accuracy, and stability. © 2007 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq, 2007     

Numerical solutions of the modified Burgers’ equation by Petrov-Galerkin method

The modified Burgers’ equation (MBE) is solved numerically by the Petrov-Galerkin method using a linear hat function as the trial function and a cubic B-spline function as the test function. Product approximation has been used in this method. A linear stability analysis of the scheme shows it to be unconditionally stable. The accuracy of the presented method is demonstrated by two test problems. The numerical results are found in good agreement with the exact solutions.     

高阶非线性薛定谔方程的精确解

通过修正的映射方法和推广的映射方法，我们得到了高阶非线性薛定谔方程新的精确解，它们是两个不同的雅可比椭圆函数的线性组合．并研究了在极限情况下高阶非线性薛定谔方程的解．     

Isolated calmness of solution mappings in convex semi-infinite optimization

This paper is concerned with isolated calmness of the solution mapping of a parameterized convex semi-infinite optimization problem subject to canonical perturbations. We provide a sufficient condition for isolated calmness of this mapping. This sufficient condition characterizes the strong uniqueness of minimizers, under the Slater constraint qualification. Moreover, on the assumption that the objective function and the constraints are linear, we show that this condition is also necessary for isolated calmness.     

A Petrov-Galerkin method for solving the generalized equal width (GEW) equation

The generalized equal width (GEW) equation is solved numerically by the Petrov-Galerkin method using a linear hat function as the test function and a quadratic B-spline function as the trial function. Product approximation has been used in this method. A linear stability analysis of the scheme shows it to be conditionally stable. Test problems including the single soliton and the interaction of solitons are used to validate the suggested method, which is found to be accurate and efficient. Finally, the Maxwellian initial condition pulse is studied.     

运筹学中若干线性目标规划和线性规划的人工智能-代数解法

运筹学中的线性目标规划和线性规划问题一直分别采用修正单纯形法和单纯形法求解.当变量稍多时计算还是有些繁琐、费时,最近作者通过研究发现,可应用人工智能-代数方法求得这两类问题的解,而且具有相当广泛的适用性.若干例题说明,本法的结果和传统方法的结果由于传统算法在计算中发生的错误,除少数例外大都是一致的.本文的一个 重要目的是希望和广大读者一起研究该方法是否具有通用性.     

On the linear convergence of the alternating direction method of multipliers

We analyze the convergence rate of the alternating direction method of multipliers (ADMM) for minimizing the sum of two or more nonsmooth convex separable functions subject to linear constraints. Previous analysis of the ADMM typically assumes that the objective function is the sum of only two convex functions defined on two separable blocks of variables even though the algorithm works well in numerical experiments for three or more blocks. Moreover, there has been no rate of convergence analysis for the ADMM without strong convexity in the objective function. In this paper we establish the global R-linear convergence of the ADMM for minimizing the sum of any number of convex separable functions, assuming that a certain error bound condition holds true and the dual stepsize is sufficiently small. Such an error bound condition is satisfied for example when the feasible set is a compact polyhedron and the objective function consists of a smooth strictly convex function composed with a linear mapping, and a nonsmooth \(\ell _1\) regularizer. This result implies the linear convergence of the ADMM for contemporary applications such as LASSO without assuming strong convexity of the objective function.     

Classical solutions of two-dimensional stokes problems on non-smooth domains. Part 2: Collocation method for the radon equation

The non-uniquely solvable Radon boundary integral equation for the two-dimensional Stokes-Dirichlet problem on a non-smooth domain is transformed into a well posed one by a suitable compact perturbation of the velocity double-layer potential operator. The solution to the modified equation is decomposed into a regular part and a finite linear combination of intrinsic singular functions whose coefficients are computed from explicit formulae. Using these formulae, the classical collocation method, defined by continuous piecewise linear vector-valued basis functions, which converges slowly because of the lack of regularity of the solution, is improved into a collocation dual singular function method with optimal rates of convergence for the solution and for the coefficients of singularities.     

应用结构异质性阈值模型定位阈值性状基因位点 : 仿真研究(英文)

蒙特卡罗仿真研究表明 :对处理在基因定位阈值性状中出现的超离中趋势现象 (结构异质性 ) ,我们提出的结构异质性模型是一个高效的统计方法 .它表现在高效率的统计检验、准确的参数估计和基因定位等方面 .病态或奇异费歇信息矩阵是在基因连锁定位分析中的一个突出的算法问题 ,仿真数据显示它们的发生率可以达到 2 8% .我们提出的应用奇异根分解方法可以有效地解决这一算法问题 .对比常规阈值模型 ,结构异质性阈值模型有较高的算法稳定性 .