小波检测并估计非参函数变点

研究随机设计下非参函数变点的小波检测与估计问题.将小波方法与设计点转化方法相结合给出变点的检测统计量并研究检测的一致性.给出了变点个数和变点位置的估计量,证明了变点个数估计量的相合性并得到变点位置估计量的收敛速度.     

0-1序列中变点的贝叶斯分析

对于0-1(伯努利)序列中的变点问题,本文提出了一个确定变点的个数和位置的贝叶斯方法。首先借助于二分法把变点个数的确定问题转化为一系列对没有变点和仅有一个变点的模型进行比较的问题,然后通过贝叶斯因子进行模型比较。本文得到了贝叶斯因子和未知变点的后验分布的显式表达式。最后,通过对上证指数数据的实证分析阐释了所提方法的有效性。     

一个基于主成分分析的探测高维变点的方法

本文介绍了一个有效的处理高维变点问题的方法。我们先将数据矩阵使用主成分分析的方法投影到低维空间,然后再利用传统变点的方法来进行估计。在变点个数未知时,我们使用交叉核实的方法来估计变点个数。在数值模拟研究中,我们将新方法同一些已有的方法进行了比较,在估计的准确度和计算时间等方面都要优于其他方法。     

潜周期模型中频率变点的估计

本文用逐段计算周期图的办法研究了带有频率交点的潜周期模型估计问题，给出了交点数目、位置和潜频率的强相合估计．数值模拟表明本文方法对变点个数和潜频率估计很好，但是要准确估计交点位置需要较大样本量，估计对于噪声水平较高情况仍然有效。     

带未知多个变点的均值突变模型的多阶段估计

本文基于局部比较法,提出了带未知多个变点的均值突变模型中变点个数与变点位置的多步估计.首先估计出可能变点位置的报警区间序列;然后根据这些报警区间初步估计出变点个数及变点位置;最后对虚假变点进行删除确定最终变点个数及变点位置.理论结果表明:几乎必然的,当n充分大时,变点个数估计将严格等于变点个数真值,所得最终报警区间将严格包含变点真值;变点估计量具有强相合性.随机模拟结果表明,多步估计在不同条件下均能准确估计变点个数和变点位置.     

无穷方差厚尾过程变点的小波估计

研究随机设计下噪声为厚尾随机变量时非参数函数中的变点估计问题.首先,通过设计变换将随机设计转化为等间距固定设计,进而利用小波方法估计变换后的变点的位置,再利用逆设计变换求得随机设计下变点位置的估计,并给出估计的收敛速度.模拟研究结果说明对于无穷方差厚尾过程中的变点估计问题小波方法是有效的.     

位置参数变点估计的强收敛速度

对至多一个变点的位置参数变点模型,文中在运用滑窗方法给出了检测变点存在性的U-统计量基础上,进一步研究了变点估计量的强相合性和强收敛速度.     

含结构变点无限方差序列的伪回归检验

基于最小二乘回归理论,研究了含结构变点无限方差序列的伪回归检验.结果表明,对两列含结构变点无限方差序列进行线性回归分析时,当两序列尾部指数之和小于1.5时,无论结构变点位置是否相同,t检验统计量均发散,导致伪回归现象的出现.究其原因,结构变点增加了回归误差的持久性,从而产生伪回归.蒙特卡罗数值模拟结果表明,伪回归现象不仅受序列尾部指数的影响,且对结构变点的位置敏感.     

正态分布方差变点的检验

本文讨论了正态分布方差只有一个变点的检验问题，我们构造了三个检验统计量，其中L检验基于非参数U统计量，B检验基于Bayes方法，R检验由极大似然比方法导出．本文给出了L、B、R检验的渐近临界值，并用MonteCarlo模拟方法研究了这三个检验与平方的CUSUM检验以及LM检验的势，并进行了比较。当变点在序列的前一半位置时，L和R检验较好，当变点在序列的后一半位置时，平方的CUSUM和B检验较好．     

变点统计分析简介(Ⅳ)局部比较法

一、方法的思想 在变点附近的“局部”中,某种量有了显著变化,它可以通过适当的估计去显示出来.在非变点附近的局部中,量保持稳定.这个差别就提供了发现变点的一个一般性方法:考察某种有针对性的统计量在各个“局部”内的变化,取其最显著之处作为变点位置的估计.当然,即使某个点非变点,但随机误差可能偶然以一种方式出现,使人误以为该处有显著变化,相反的情况也存在.这就需要通过检验来判定,那种程度的变化可视为显著的或随机性的,这涉及复杂的分布问题. 这个方法可用于种种模型,但在本文中,我们将只介绍其在一个简单模型──均值变点模型下…     

Local Likelihood Method in the Problems Related to Chang Points

In this paper, the so-called local likelihood method is suggested for solving the change point problems when the data are distributed as multivariate normal. The detection procedures proposed not only provide strongly consistent estimates for the number and locations of the change points, but also simplify significantly the computation.     

非参数回归中方差变点的小波检测(英文)

本文主要研究了非参数回归模型中方差函数的变点, 利用小波方法构造的检验量来检测方差中的变点,建立了这些检验量的渐近分布, 并且运用这些检验量构造了方差变点的位置和跳跃幅度的估计, 给出了这些估计的渐近性质, 并进一步通过随机模拟验证了本文方法在有限样本下的性质.     

On detection of change points using mean vectors

In this paper, the authors consider the problem of change points within the framework of model selection and propose a procedure for estimating the locations of change points when the number of change points is known. The strong consistency of this procedure is also established. The problem of detecting change points is discussed within the framework of the simultaneous test procedure. The case where the number of change points is unknown will be discussed in another paper.This project is supported by the National Natural Science Foundation of China and by the Air Office of Scientific Research of the United States.     

Reduction in mean residual life in the presence of a constant competing risk

The addition of a constant ‘competing risk’ corresponding to an additional, usually less significant, source of failure, frequently improves the fit in reliability and survival analysis. This is often termed a ‘lift’, as the effect is to increase the hazard rate (HR) function by a constant, which does not, of course, change the shape and hence the turning points of the HR function. However, lifting the HR function does not, in general, mean lowering the corresponding mean residual life (MRL) function by a constant, and so the MRL turning points, unlike those of the HR function are not invariant. The MRL turning points are used in, for example, defining burn‐in procedures in reliability engineering, and determining premiums in insurance. Hence, it is of interest to examine the changes in the shape of the MRL function, and in the locations of its turning points, resulting from a lift in the HR function. We discuss these problems in detail, with reference to a number of common distributions in reliability and mortality modeling. Copyright © 2007 John Wiley & Sons, Ltd.     

A comprehensive study of optimization algorithm for wireless coverage in indoor area

Coverage optimization using minimum number of transmitters is critical to service providers and vendors that need to control the coverage as well as the huge costs involved. In this regards, an existing coverage algorithm for determining the minimum number of transmitting antennas as well as their appropriate locations to provide the optimized wireless coverage for indoor environment is studied in this paper. The algorithm uses ray-tracing to predict the signal distribution from the transmitter to the sampling points (receivers) and genetic algorithm to determine the minimum number of transmitters and their corresponding locations to achieve the optimum wireless coverage. The complexity and performance of the algorithm are also analyzed and it is found that it has exponentially increasing complexity of $2^{n}$ and the change of computation time is greater with small change of the number of receiving points. Moreover, under the multi-transmitter scenario (real case), the accuracy achieved by fading, coverage ability, and signal to noise ratio is in the range of 96–99 %.     

Modeling short‐term post‐offering price–volume relationships using Bayesian change‐point panel quantile regression

We propose a special panel quantile regression model with multiple stochastic change‐points to analyze latent structural breaks in the short‐term post‐offering price–volume relationships in China's growth enterprise market where the piecewise quantile equations are defined by change point indication functions. We also develop a new Bayesian inference and Markov chain Monte Carlo simulation approach to estimate the parameters, including the locations of change points, and put forth simulation‐based posterior Bayesian factor tests to find the best number of change points. Our empirical evidence suggests that the single change point effect is significant on quantile‐based price–volume relationships in China's growth enterprise market. The lagged initial public offering (IPO) return and the IPO volume rate of change have positive impacts on the current IPO return before and after the change point. Along with investors' gradually declining hot sentiment toward a new IPO, the market index volume rate of change induces the abnormal short‐term post‐offering IPO return to move back to the equilibrium. Copyright © 2015 John Wiley & Sons, Ltd.     

Enumeration and Search Procedures for a Hub Location Problem with Economies of Scale

Within a communications or transportation network, in which a number of locations exchange material or information, hubs can be used as intermediate switching points. In this way, traffic can be consolidated on inter-hub links and, thus, achieve economies of scale in transport costs. Recently, O'Kelly and Brian in 1998 proposed a model (termed the FLOWLOC model) that treats these economies of scale by means of piecewise-linear concave cost functions on the interhub arcs. We show that, for a fixed set of hubs, the FLOWLOC model can be solved using the classic Uncapacitated Facility Location Problem (UFLP). This observation then motivates an optimal enumeration procedure for the FLOWLOC model, as well as some search heuristics that are based upon tabu search and greedy random adaptive search procedures (GRASP). These search procedures would be especially applicable for large-sized problems. Some computational experience is described.     

Optimal locations for a class of nonlinear, single-facility location problems on a network

This paper investigates a class of single-facility location problems on an arbitrary network. Necessary and sufficient conditions are obtained for characterizing locally optimal locations with respect to a certain nonlinear objective function. This approach produces a number of new results for locating a facility on an arbitrary network, and in addition it unifies several known results for the special case of tree networks. It also suggests algorithmic procedures for obtaining such optimal locations.     

Optimal location of dwell points in a single loop AGV system with time restrictions on vehicle availability

Since the workload of a manufacturing system changes over time, the material handling equipment used in the facility will be idle at certain time intervals to avoid system overload. In this context, a relevant control problem in operating an automated guided vehicle (AGV) system is where to locate idle vehicles. These locations, called dwell points, establish the response times for AVG requests. In this article, a dynamic programming algorithm to solve idle vehicle positioning problems in unidirectional single loop systems is developed to minimize the maximum response time considering restrictions on vehicle time available to travel and load/unload requests. This polynomial time algorithm finds optimal dwell points when all requests from a given pick-up station are handled by a single AGV. The proposed algorithm is used to study the change in maximum response time as a function of the number of vehicles in the system.     

Detecting change points in the stress‐strength reliability P(X < Y)

We address the statistical problem of detecting change points in the stress‐strength reliability R=P(X<Y) in a sequence of paired variables (X,Y). Without specifying their underlying distributions, we embed this nonparametric problem into a parametric framework and apply the maximum likelihood method via a dynamic programming approach to determine the locations of the change points in R. Under some mild conditions, we show the consistency and asymptotic properties of the procedure to locate the change points. Simulation experiments reveal that, in comparison with existing parametric and nonparametric change‐point detection methods, our proposed method performs well in detecting both single and multiple change points in R in terms of the accuracy of the location estimation and the computation time. Applications to real data demonstrate the usefulness of our proposed methodology for detecting the change points in the stress‐strength reliability R. Supplementary materials are available online.