A perfect sampling method for exponential family random graph models

Generation of deviates from random graph models with nontrivial edge dependence is an increasingly important problem. Here, we introduce a method which allows perfect sampling from random graph models in exponential family form (“exponential family random graph” models), using a variant of Coupling From The Past. We illustrate the use of the method via an application to the Markov graphs, a family that has been the subject of considerable research. We also show how the method can be applied to a variant of the biased net models, which are not exponentially parameterized.     

A Classification of Reproducible Natural Exponential Families in the Broad Sense

Let be a positive Radon measure on having a Laplace transform L and F=F() be a natural exponential family (NEF) generated by . A positive Radon measure with >0 and its associated NEF F =F( ) are called the th convolution powers of and F, respectively, if . Let f _, (F) be the image of an NEF F under the affine transformation f _, :xx+. If for a given NEF F, there exists a triple (,,) in such that f _, (F)=F , we call F a reproducible NEF in the broad sense. In other words, an NEF F is reproducible in the broad sense if a convolution power of F equals an affine transformation of F. Clearly, for =1, F is reproducible in the broad sense if there exists an affinity under which F is invariant. In this paper we obtain a complete classification of the class of reproducible NEF's in the broad sense. We show that this class is composed of infinitely divisible NEF's and that it contains the class of NEF's having exponential and power variance functions as well as NEF's constituting discrete versions of the latter NEF's. We also provide a characterization of the reproducible NEF's in the broad sense in terms of their associated exponential dispersion models.     

一类非线性离散系统的指数二分性及其在数值分析和计算中的应用

本文中我们考虑一类二阶非线性常微分方程的边值问题的迎风差分格式.我们运用奇异摄动方法构造了该迎风差分方程解的渐近近似,并利用指数二分性理论证明了有一个低阶方程其解是该迎风方程式的在边界外的一个良好近似.我们还构造了校正项,使校正项与低阶方程的解之和是一个渐近近似.最后一些数值例子用于显示本文方法的应用.     

关闭—启动型Geom／G／1离散排队及其在ATM网络中的应用

研究带有关闭延迟和启动时间的Geom/G/1离散时间排队,导出了稳态队长、等待时间的分布及其随机分解结果,该模型可用于ATM网络的虚通道分析,给出响应时间、启动率,闲置率等指标公式。     

基于MCMC稳态模拟的指数回归模型及其应用

讨论了加速失效模型族中最简单而又十分重要的指数回归模型，利用贝叶斯方法提高了该模型的有效性。为了较好的解决高维数值积分在实际应用中的难题，提出了对寿命服从指数分布的产品，运用基于Gibbs抽样的马尔科夫链蒙特卡洛（Markov Chain Monte Carlo，MCMC）方法动态模拟出参数后验分布的马尔科夫链，在回归参数的先验分布为多元正态分布时，给出随机截尾条件下，回归参数在指数回归模型中的贝叶斯估计，提高了计算的精度。借助数据仿真分析说明了利用WinBUGS（Bayesian inference Using Gibbs Sampling）软件包进行建模分析的过程，证明了该模型在可靠性应用中的直观性与有效性。     

一类嵌套多元指数分布相关参数的矩型估计

本文研究一类具有三个相关参数的嵌套多元指数分布,其生存函数为其中,0相似文献   

指数分布抽样基本定理及在三参数一般指数分布参数估计中的应用

讨论三参数一般指数分布的参数估计,首先讨论了三参数一般指数分布参数的最大似然估计的求解问题,当其中参数α=1时,应用指数分布抽样基本定理,得到了三参数一般指数分布其它参数的一致最小方差无偏估计;并且由此给出求解三参数一般指数分布参数最大似然估计的迭代方法,得到了三参数一般指数分布参数最大似然估计的近似值,给出了模拟结果以说明迭代方法的收敛性;并以相关文献的观察数据作为样本,得到了三参数一般指数分布的参数估计,从而说明了迭代方法的有效性.     

Two-Stage Nested Partitions Method for Stochastic Optimization

We address the problem of optimizing over a large but finite set when the objective function does not have an analytical expression and is evaluated using noisy estimation. Building on the recently proposed nested partitions method for stochastic optimization, we develop a new approach that combines this random search method and statistical selection for guiding the search. We prove asymptotic convergence and analyze the finite time behavior of the new approach. We also report extensive numerical results to illustrate the benefits of the new approach.     

A General Version of the Retract Method for Discrete Equations

We study a problem concerning the compulsory behavior of the solutions of systems of discrete equations u（k ＋ 1） = F（k, u（k））, k ∈ N（a） = {a, a ＋ 1, a ＋ 2 }, a ∈ N,N= {0, 1,... } and F ： N（a） × R^n→R^n. A general principle for the existence of at least one solution with graph staying for every k ∈ N（a） in a previously prescribed domain is formulated. Such solutions are defined by means of the corresponding initial data and their existence is proved by means of retract type approach. For the development of this approach a notion of egress type points lying on the defined boundary of a given domain and with respect to the system considered is utilized. Unlike previous investigations, the boundary can contain points which are not points of egress type, too. Examples are inserted to illustrate the obtained result.     

连续型随机向量联合熵的离散方差分离估计

提出了一种"离散方差分离"法,用于连续型随机向量联合熵的估计.方法分为"方差分离"和"离散"两个步骤.前者通过分离"标准熵"与"标准差对数和"来避免维数灾害;后者通过各分量的"最佳分割数"来离散连续型随机向量,从而避开了联合密度估计.仿真实验表明:该方法以很低的计算复杂度,准确地逼近了理论值.     

应力服从一类嵌套多元指数分布的结构可靠度估计

本文研究应力服从一类嵌套多元指数分布, 强度服从指数分布的应力---强度结构可靠度模型.分别在强度参数未知、应力参数已知和强度参数已知、应力参数未知的情况下给出了结构可靠度$P_{A}$的估计$\wh{P}_{A1}$和$\wh{P}_{A2}$,并讨论了它们的渐近性质,而且获得了$P_{A}$的近似置信区间.最后对这两种情况下模型结构可靠度的估计$\wh{P}_{A1}$和$\wh{P}_{A2}$进行了随机模拟,随机模拟结果令人满意.     

双边定时截尾样本下广义逆指数分布形状参数的估计

先给出了广义逆指数分布在双边定时截尾样本下形状参数的最大似然估计,并不能得到估计的显式表达式,但证明了参数在(0,+∞)上最大似然估计是唯一存在的.其次提出用EM算法求出形状参数的估计且该估计具有良好的收敛性,还给出了形状参数的EM估计的渐近方差和近似置信区间;最后通过数值模拟,对形状参数的最大似然估计和EM估计的效果...     

离散投资组合问题的一种基于Bundle对偶搜索的精确算法

本文提出了离散均值-方差投资组合模型的一种新的精确算法.该算法是一个基于拉格朗日松弛和Bundle对偶搜索的分枝定界算法.我们分别用随机产生的数据和美国股票市场的真实数据进行了数值实验,并与传统次梯度对偶搜索进行了比较,数值结果表明本文提出的算法对解决中小规模的离散投资组合问题是有效的.     

A Product Theorem for Markov Chains with Application to PF-Queueing Networks

Queueing networks in random environments represent more realistic models of computer and telecommunication systems than classical product form networks. This is due to the fact that network behaviour often depends on human activities which may vary according to daytime dependent behavioural patterns as well as physiological and mental indexes. In this paper we establish a product connection theorem for Markov chains which contains some corresponding results for spatial processes as well as for queueing networks in random environment as special cases. We demonstrate how our results can be applied to construct an adequate model for wireless networks with hook up capacity.     

离散单因素投资组合模型的对偶算法

本文研究金融优化中的离散单因素投资组合问题,该问题与传统投资组合模型的不同之处是决策变量为整数(交易手数),从而导致要求解一个二次整数规划问题．针对该模型的可分离性结构,我们提出了一种基于拉格朗日对偶和连续松弛的分枝定界算法。我们分别用美国股票市场的交易数据和随机产生的数据对算法进行了测试．数值结果表明该算法是有效的,可以求解多达150个风险证券的离散投资组合问题．     

关于算术、对数、指数、几何平均不等式的概率证法

本文用概率方法证明了关于两个相异的正数的几何平均、对数平均、指数平均和算术平均的不等式关系 .     

多变量、多约束连续或离散的非线性规划的一个通用算法

利用目标函数对约束函数关于设计变量的一阶微分或差分之比,给出了一个求解非线性规划的通用算法．不论变量和约束有多少,也不论变量是连续的还是离散的,这一算法都比较有效,尤其对离散非线性规划更有效．该方法是一种搜索法,勿需解任何数学方程,只需要计算函数值以及函数对变量的偏微分或差分值．许多数值例题和运筹学中一些经典问题,如1) 一、二维的背包问题;2) 一、二维资源分配问题;3) 复合系统工作可靠性问题;4) 机器负荷问题等,经用此法求解验证均较传统方法更有效和可靠．该方法的主要优点是:1) 不受问题的规模限制;2) 只要在可行域（集）内存在目标函数和约束函数及其一阶导数或差分的值,肯定可以搜索到最优的解,没有不收敛和不稳定的问题．     

奇异差分边值问题上下解方法的推广

This paper presents new existence results for singular discrete boundary value problems. In particular our nonlinearity may be singular in its dependent variable and is allowed to change sign.     

Identification of Regeneration Times in MCMC Simulation,With Application to Adaptive Schemes

Regeneration is a useful tool in Markov chain Monte Carlo simulation because it can be used to side-step the burn-in problem and to construct better estimates of the variance of parameter estimates themselves. It also provides a simple way to introduce adaptive behavior into a Markov chain, and to use parallel processors to build a single chain. Regeneration is often difficult to take advantage of because, for most chains, no recurrent proper atom exists, and it is not always easy to use Nummelin's splitting method to identify regeneration times. This article describes a constructive method for generating a Markov chain with a specified target distribution and identifying regeneration times. As a special case of the method, an algorithm which can be “wrapped” around an existing Markov transition kernel is given. In addition, a specific rule for adapting the transition kernel at regeneration times is introduced, which gradually replaces the original transition kernel with an independence-sampling Metropolis-Hastings kernel using a mixture normal approximation to the target density as its proposal density. Computational gains for the regenerative adaptive algorithm are demonstrated in examples.     

循环搜索法在三参数双指数曲面拟合中的应用

在最小二乘意义下用 Gauss-Newton法拟合三参数双指数曲面z =γ[exp( -αx) +exp( -βy) ]时 ,充分利用观测值 ,应用循环搜索法确定了参数初始值 .并成功地以由 Meyer &Roth给出的观测值为例对此进行了验证 .