基于Koopman-Darmois分布族的序贯网图检验方法

通常高成本、破坏性的抽样检验需采用具有最大样本量限制的序贯检验方法．立足于设计出最大样本量尽量小的序贯检验方案,本文基于Koopman-Darmois分布族建立了序贯网图检验方法．与目前广泛采用的截尾序贯概率比检验方法相比,序贯网图检验方案具有更小的样本量上界,更适合高成本、破坏性的抽样检验．     

计数型最小样本量截尾值的序贯检验

对具有"高成本、破坏性"计数型产品的可靠性验收试验,序贯检验方案的样本量截尾值对试验的成本预算起着决定性作用.为降低试验的成本预算,本文对最小样本量截尾值序贯检验(minimum truncated sample size sequential test, MTST)进行研究,给出MTST的定义、性质及求解方法.通过与目前广泛采用的国际标准IEC1123及序贯网图检验进行比较,结果表明MTST显著地减少了样本量截尾值.为进一步减少序贯检验的样本量截尾值,当已知产品质量的先验信息时,本文研究了Bayes最小样本量截尾值序贯检验方案(Bayesian minimum truncated sample size sequential test, BMTST),与MTST的比较表明, BMTST极大地减少了序贯检验的样本量截尾值及平均试验次数,能更好地缩减"高成本、破坏性"产品的试验成本预算及平均试验费用.     

关于序贯检验的Bayes分析

本文将充分利用验前信息的Ｂａｙｅｓ方法与序贯检验相结合,推导出了Ｂａｙｅｓ序贯检验方法。文章从表面形式上的决策方案推导到实质上真正的决策方案;并对飞行器落点密集度进行了模拟计算的具体分析,最后得出结论,Ｂａｙｅｓ序贯检验方法对具有昂贵成本的产品试验和我国航天试验有较大的应用价值．     

计数型二次序贯网图检验

序贯概率比检验(SPRT)是应用非常广泛的抽样检验方法, 序贯网图检验在控制最大样本量方面很好地改进了SPRT, 但其结果还有进一步改进的余地, 为此, 我们建立了二次序贯网图检验, 计算结果表明, 它比原先的计数序贯网图检验有更好的效果.     

序贯Lq似然比型检验

本文研究了序贯Lq似然比型检验(简写为SLqRT),以及SLqRT的一些性质.基于其中一个性质-基本恒等式,得到SLqRT的势函数和平均子样容量.为了节约成本,通过最小化平均子样容量来选择合适的q值,从而进行指数分布的假设检验.此外,给出了理论推导及两个模拟算例来说明SLqRT的稳健性,并计算了不同污染参数下粗差模型的序贯概率比检验(简写为SPRT)和SLqRT的平均子样容量.结果表明:在平均子样容量下,SLqRT的效果优于SPRT.     

越界概率与序贯似然比检验简介

一、引言二次大战时期,为了提高军火试验中抽样调查的效率,A.Wald等人首先较为系统地研究了序贯分析。Wald的奠基性著作问世以来,序贯分析的研究范围越来越大,理论和应用都有了很大的发展。尤其是在医学诊断中的应用有力地促进了序贯的理论研究。本文从一个侧面反映部分序贯分析的新动向。作重介绍从七十年代中期开始受到广泛注意的一种序贯似然比检验方法——重复显著性检验(Repeated Significance Test)。这种检验与在统计中得到广泛应用的越界概率(Boundary Crossing Probability)问题密切相关。本文试图总结有关重复显著性检验的最新成果;并扼要阐述以统计应用为目的的求越界概率的几种常用方法;我们还介绍一种与越界概率有关的统计模型——转折点(Change-Point)问题。1986年D.Siegmund发表于《Annals of Statistics》上的特邀论     

几何分布的一类贝叶斯停止判决法则

1 引言 以节约样本为目的序贯方法在数理统计中占有重要地位.从实际工作的角度出发,人们往往更强调时间的价值,希望当有足够的证据做出推断时应尽早停止试验,这样就提出了时间序贯计划.近年来,时间序贯方法得到了迅速发展(见[1—6]).[3]和[6]讨论了指数分布的时间序贯检验问题. [5]讨论了单试验平台情形,几何分布的时间序贯检验问题,适合于受试样品比较昂贵的情形.本文讨论多试验平台,受试品比较廉价而试验时间(次数)比较宝贵情形的几何分布的检验问题.     

面板数据下序贯模型变点的渐近检验法

本文研究了面板数据模型下变点是否存在的检验问题.对于面板序贯模型下的变点,利用构造的检验统计量及其极限分布的方法,获得了关于变点的一种渐近检验法.随后进行Monte-Carlo数值模拟,在模拟中对检验方法的经验检验水平、检验功效以及变点后的停时三个判断标准进行了考察.结果显示无论在理论还是数值模拟上,我们提出的渐近检验法均表现优良.推广了现有文献的关于单序列变点的序贯检验方法.     

序贯k-out-of-n系统在序约束下参数的估计和算法

本文主要研究带有协变量的序贯k-out-of-n模型.我们假定给定协变量寿命的分布是指数分布,对指数分布的刻度参数建立了对数线性模型.研究了在序约束下模型参数的最大似然估计及最大似然估计量的性质,并且给出了最大似然估计的具体算法并进行了模拟.     

累积Logistic回归模型结构变点的序贯检验

分类时间序列在生物医学、社会学和遗传学等领域有着广泛的应用,累积Logistic回归模型是分类时间序列建模的一类重要模型.本文基于偏似然得分过程(Partial likelihood score process)提出一种变点序贯检验方法,监测累积Logistic回归模型的结构是否发生变化.原假设下推导检验统计量的极限分...     

A note on the Bayes factor in a semiparametric regression model

In this paper, we consider a semiparametric regression model where the unknown regression function is the sum of parametric and nonparametric parts. The parametric part is a finite-dimensional multiple regression function whereas the nonparametric part is represented by an infinite series of orthogonal basis. In this model, we investigate the large sample property of the Bayes factor for testing the parametric null model against the semiparametric alternative model. Under some conditions on the prior and design matrix, we identify the analytic form of the Bayes factor and show that the Bayes factor is consistent, i.e. converges to infinity in probability under the parametric null model, while converges to zero under the semiparametric alternative, as the sample size increases.     

The theory of parametric oscillations

The correctness of the existing definitions of the parametric oscillations of linear and non-linear systems is discussed. The possibility of an erroneous choice of the mathematical parametric model instead of the autooscillatory model, connected with the existence in such systems of the same periodic solutions, is pointed out. Some non-local properties of parametric oscillations in Hamiltonian systems are established. It is shown, in particular, that stability regions are convex with respect to the frequency of the parametric excitation (i.e., all the points between the boundaries of neighbouring instability regions correspond to stable solutions). At the critical frequencies of parametric resonance the well-known Rayleigh and Zhuravlev theorems on the behaviour of the frequencies of natural oscillations when the stiffness and inertia changes are generalized. Some additional assertions on the limits of the first instability region for the Hill vector equations are established.     

Realtime control of robots with initial value perturbations via nonlinear programming methods

The mathematical model of an industrial robot with initial value perturbations is considered as a parametric nonlinear control problem subject to control and state constraints. Based on recent stability results for parametric control problems, a robust nonlinear programming method is proposed for computing the sensitivity derivatives of optimal solutions. Real-time control approximations of perturbed optimal solutions are obtained by evaluating a first order Taylor expansion of the perturbed solution. The efficiency of the real-time approximation is demonstrated for the robot model     

Interpolatory model reduction of parameterized bilinear dynamical systems

Interpolatory projection methods for model reduction of nonparametric linear dynamical systems have been successfully extended to nonparametric bilinear dynamical systems. However, this has not yet occurred for parametric bilinear systems. In this work, we aim to close this gap by providing a natural extension of interpolatory projections to model reduction of parametric bilinear dynamical systems. We introduce necessary conditions that the projection subspaces must satisfy to obtain parametric tangential interpolation of each subsystem transfer function. These conditions also guarantee that the parameter sensitivities (Jacobian) of each subsystem transfer function are matched tangentially by those of the corresponding reduced-order model transfer function. Similarly, we obtain conditions for interpolating the parameter Hessian of the transfer function by including additional vectors in the projection subspaces. As in the parametric linear case, the basis construction for two-sided projections does not require computing the Jacobian or the Hessian.     

部分线性模型基于参数信息的统计推断

针对部分线性模型提出了一种新的估计方法-Profile局部最小二乘估计,方法结合了非参数部分的参数信息.另外对于部分线性模型中非参数部分是否为某一参数函数的检验问题,基于比较原假设与备择假设下模型拟合的残差平方和的思想构造了检验统计量,并给出了计算检验p-值的精确方法和三阶矩χ2逼近方法.     

Dynamical stability of a viscoelastic bar

In the paper the dynamic stability of a simply supported viscoelastic bar of the Zener material model is investigated. The bar is subjected to a parametric excitation of a periodic nature. The physical model of the system is of the continuous type. However, the proposed approach yields a transformation of the mathematical model from the partial differential equation to an ordinary one. Such a transformation is made possible by the use of an approximation function for the bending line of the bar. This way, the system is governed by a homogeneous, ordinary differential equation of the third order with periodic coefficients. For stability studies the Floquet theory is applied. A sensitivity analysis of a parametric periodic system is discussed, i.e. the influence of stiffness and damping coefficient of the Zener model on stability of the differential equation that describes the vibration of the viscoelastic bar. Furthermore the stabilization process of an unstable parametric system was realized. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

On the explosive instability in a three‐species food chain model with modified Holling type IV functional response

In earlier literature, a version of a classical three‐species food chain model, with modified Holling type IV functional response, is proposed. Results on the global boundedness of solutions to the model system under certain parametric restrictions are derived, and chaotic dynamics is shown. We prove that in fact the model possesses explosive instability, and solutions can explode/blow up in finite time, for certain initial conditions, even under the parametric restrictions of the literature. Furthermore, we derive the Hopf bifurcation criterion, route to chaos, and Turing bifurcation in case of the spatially explicit model. Lastly, we propose, analyze, and simulate a version of the model, incorporating gestation effect, via an appropriate time delay. The delayed model is shown to possess globally bounded solutions, for any initial condition. Copyright © 2017 John Wiley & Sons, Ltd.     

Asymptotic Normality of a Combined Regression Estimator

In this paper, we propose a combined regression estimator by using a parametric estimator and a nonparametric estimator of the regression function. The asymptotic distribution of this estimator is obtained for cases where the parametric regression model is correct, incorrect, and approximately correct. These distributional results imply that the combined estimator is superior to the kernel estimator in the sense that it can never do worse than the kernel estimator in terms of convergence rate and it has the same convergence rate as the parametric estimator in the case where the parametric model is correct. Unlike the parametric estimator, the combined estimator is robust to model misspecification. In addition, we also establish the asymptotic distribution of the estimator of the weight given to the parametric estimator in constructing the combined estimator. This can be used to construct consistent tests for the parametric regression model used to form the combined estimator.     

Modeling framework for optimal evacuation of large-scale crowded pedestrian facilities

The paper presents a simulation–optimization modeling framework for the evacuation of large-scale pedestrian facilities with multiple exit gates. The framework integrates a genetic algorithm (GA) and a microscopic pedestrian simulation–assignment model. The GA searches for the optimal evacuation plan, while the simulation model guides the search through evaluating the quality of the generated evacuation plans. Evacuees are assumed to receive evacuation instructions in terms of the optimal exit gates and evacuation start times. The framework is applied to develop an optimal evacuation plan for a hypothetical crowded exhibition hall. The obtained results show that the model converges to a superior optimal evacuation plan within an acceptable number of iterations. In addition, the obtained evacuation plan outperforms conventional plans that implement nearest-gate immediate evacuation strategies.     

A conservative estimator for the proportion of false nulls based on Dvoretzky, Kiefer and Wolfowitz inequality

We propose a simple estimator for the weight in a two-component mixture model between a known and an unknown density. We make no parametric assumptions about the unknown component and estimate the weight conservatively, i.e., the estimate is smaller than the true value with high probability. A brief simulation is used to compare the proposal with already available conservative approaches.