扩散先验分布下Bayes多总体分类判别方法的构造

本文研究了各总体服从多元正态分布 ,其未知参数的先验分布均为扩散先验分布时 ,如何利用待判样品的预报密度函数、构造后验概率比并据此对样品进行分类与判别 ;此方法并不需要假设各总体分布的协方差相同 ,而且在预试样本容量较小时仍然可行。     

A Nussbaum gain adaptive synchronization of a new hyperchaotic system with input uncertainties and unknown parameters

The sign of unknown input coefficients is assumed to be known in most papers about the input uncertainties. In this paper, a Nussbaum gain method is adopted to cope with the situation that both the sign and the value of input are unknown. And the unknown parameters can be estimated under the situation of unknown sign of control. The synchronization is achieved for a class of hyperchaotic systems with unknown parameters and input uncertainties by adopting of the Nussbaum gain method and the global terminal adaptive method. And the conclusions are made as follows: First, the proposed method is effective in the situation that the sign of input is unknown. Second, the estimation of unknown parameters can be achieved only when the number of unknown parameters satisfied some condition and no uncertainty exist in the input of systems. Third, the unknown parameters cannot be estimated correctly with common adaptive method when there are input uncertainties in the system. But the Nussbaum gain method can get good result in the estimation of unknown parameters. At last, numerical simulations are done to show the effectiveness of the proposed method.     

Estimation,model discrimination,and experimental design for implicitly given nonlinear models of enzyme catalyzed chemical reactions

Many nonlinear models as e.g. models of chemical reactions are described by systems of differential equations which have no explicit solution. In such cases the statistical analysis is much more complicated than for nonlinear models with explicitly given response functions. Numerical approaches need to be applied in place of explicit solutions. This paper describes how the analysis can be done when the response function is only implicitly given by differential equations. It is shown how the unknown parameters can be estimated and how these estimations can be applied for model discrimination and for deriving optimal designs for future research. The methods are demonstrated with a chemical reaction catalyzed by the enzyme Benzaldehyde lyase.     

Evaluation of Branched Continued Fractions using Block-Tridiagonal Linear Systems

The convergent of an ordinary continued fraction can be computedby solving a tridiagonal linear system for its first unknown.In this paper, this approach is generalized to branched continuedfractions, and it is shown how the convergent of a branchedcontinued fraction can be considered as the first unknown ofa block-tridiagonal linear system. Hence algorithms for thesolution of such systems of equations can be used for the computationof convergents of branched continued fractions, which have applicationsin approximation theory, systems theory, etc. In future research,special attention will be paid to the use of parallel algorithms.     

Delete-group Jackknife Estimate in Partially Linear Regression Models with Heteroscedasticity

Abstract Consider a partially linear regression model with an unknown vector parameter β,an unknownfunction g(.),and unknown heteroscedastic error variances.Chen,You proposed a semiparametric generalizedleast squares estimator(SGLSE)for β,which takes the heteroscedasticity into account to increase efficiency.Forinference based on this SGLSE,it is necessary to construct a consistent estimator for its asymptotic covariancematrix.However,when there exists within-group correlation, the traditional delta method and the delete-1jackknife estimation fail to offer such a consistent estimator.In this paper, by deleting grouped partial residualsa delete-group jackknife method is examined.It is shown that the delete-group jackknife method indeed canprovide a consistent estimator for the asymptotic covariance matrix in the presence of within-group correlations.This result is an extension of that in[21].     

A Chebyshev polynomial method for line integrals with singularities

In this paper we consider a Chebyshev polynomial method for the calculation of line integrals along curves with Cauchy principal value or Hadamard finite part singularities. The major point we address is how to reconstruct the value of the integral when the parametrization of the curve is unknown and only empirical data are available at some discrete set of nodes. We replace the curve by a near‐minimax parametric polynomial approximation, and express the integrand by means of a sum of Chebyshev polynomials. We make use of a mapping property of the Hadamard finite part operator to calculate the value of the integral. This revised version was published online in June 2006 with corrections to the Cover Date.     

Knowledge discovery using a neural network simultaneous optimization algorithm on a real world classification problem

Artificial neural networks have been shown to perform well for two-group classification problems. However, current research has yet to determine a method for identifying relevant input variables in the neural network model for real world classification problems. The common practice in neural network research is to include all available input variables that could possibly contribute to the model without determination of whether they help in estimating the unknown function. One problem with this avenue of neural network research is the inability to extract the knowledge that could be useful to researchers by identifying those input variables that contribute to estimating the true underlying function of the data. A method has been proposed in past research, the Neural Network Simultaneous Optimization Algorithm (NNSOA), which was shown to be successful for a limited number of continuous problems. This research proposes using the NNSOA on a real world classification problem that not only finds good solutions for estimating unknown functions, but can also correctly identify those variables that contribute to the model.     

End-to-end availability-dependent pricing of network services

We discuss how a new pricing scheme can be integrated within a communication network. The pricing scheme is based on the availability of end-to-end communications, and is an alternative to congestion pricing, which is not applicable when communication capacity is higher than demand (as happens in most communication backbone networks). We also investigate how, based on this scheme, an optimization algorithm for updating the network topology can be applied. The network update problem is modeled as a combinatorial optimization problem, which is approximately solved using a Genetic Algorithm. The good results obtained in a case study show that the method is robust and can be applied even when end-to-end availability measures can only be computed approximately (in this case, using a Monte Carlo method). This research is part of the PAIR associated research project, supported by the INRIA, France, and has also received the support of ECOS-Sud, under Action U03E02. The participation of Pablo Rodríguez was supported by the French Embassy in Uruguay as part of the French Ministère des Affaires étrangères scientific cooperation program; and by the “Programa de Jóvenes Investigadores” of CSIC, UDELAR, Uruguay.     

Exponential Modeling with Unknown Model Order Using Structured Nonlinear Total Least Norm

A new algorithm called Structured Nonlinear Total Least Norm (SNTLN) has recently been developed for obtaining an approximate solution to the structured overdetermined nonlinear system. Both theoretical justification and computational testing show that SNTLN is an efficient method for solving structured overdetermined systems. In this paper, we present a method based on SNTLN for estimating the parameters of exponentially damped sinusoidal signals in noise when the model order is unknown. It is compared to two other existing methods to show its robustness in recovering correct values of parameters when the model order is unknown, in spite of some large errors in the measured data.     

Tail asymptotics for a generalized two-demand queueing model—a kernel method

In this paper, we consider a generalized two-demand queueing model, the same model studied in Wright (Adv. Appl. Prob., 24, 986–1007, 1992). Using this model, we show how the kernel method can be applied to a two-dimensional queueing system for exact tail asymptotics in the stationary joint distribution and also in the two marginal distributions. We demonstrate in detail how to locate the dominant singularity and how to determine the detailed behavior of the unknown generating function around the dominant singularity for a bivariate kernel, which is much more challenging than the analysis for a one-dimensional kernel. This information is the key for characterizing exact tail asymptotics in terms of asymptotic analysis theory. This approach does not require a determination or presentation of the unknown generating function(s).     

Assessment and propagation of input uncertainty in tree‐based option pricing models

This paper aims to provide a practical example of assessment and propagation of input uncertainty for option pricing when using tree‐based methods. Input uncertainty is propagated into output uncertainty, reflecting that option prices are as unknown as the inputs they are based on. Option pricing formulas are tools whose validity is conditional not only on how close the model represents reality, but also on the quality of the inputs they use, and those inputs are usually not observable. We show three different approaches to integrating out the model nuisance parameters and show how this translates into model uncertainty in the tree model space for the theoretical option prices. We compare our method with classical calibration‐based results assuming that there is no options market established and no statistical model linking inputs and outputs. These methods can be applied to pricing of instruments for which there is no options market, as well as a methodological tool to account for parameter and model uncertainty in theoretical option pricing. Copyright © 2008 John Wiley & Sons, Ltd.     

Bootstrap bandwidth selection in kernel density estimation from a contaminated sample

In this paper we consider kernel estimation of a density when the data are contaminated by random noise. More specifically we deal with the problem of how to choose the bandwidth parameter in practice. A theoretical optimal bandwidth is defined as the minimizer of the mean integrated squared error. We propose a bootstrap procedure to estimate this optimal bandwidth, and show its consistency. These results remain valid for the case of no measurement error, and hence also summarize part of the theory of bootstrap bandwidth selection in ordinary kernel density estimation. The finite sample performance of the proposed bootstrap selection procedure is demonstrated with a simulation study. An application to a real data example illustrates the use of the method. This research was supported by ‘Projet d’Actions de Recherche Concertées’ (No. 98/03-217) from the Belgian government. Financial support from the IAP research network nr P5/24 of the Belgian State (Federal Office for Scientific, Technical and Cultural Affairs) is also gratefully acknowledged.     

计算主梁绝对最大挠度的数学模型与0.618法

纵横梁桥面系统中主梁的跨度较长,纵梁上直接承受的车队荷载数量多且数值大,主梁设计中绝对最大挠度的确定是关键内容.研究了一组平行车队荷载直接沿着纵梁移动时,主梁承受结点活载下绝对最大挠度数学模型的建立;并给出了相应的计算方法,以计算机为工具,适用于任意有限多个平行移动荷载作用工况,对于主梁的设计计算与安全评估,有一定的实用价值.     

A sequential algorithm and error sensitivity analysis for the inverse heat conduction problems with multiple heat sources

This paper proposes a sequential approach to determine the unknown parameters for inverse heat conduction problems which have multiple time-dependent heat sources. There are two main aims in this study, one is to derive an inverse algorithm that can estimate the unknown conditions effectively, and the other is to bring up a theoretical sensitivity analysis to discuss what causes the growth of errors. This paper has three major achievements with regard to the literature on IHCPs, as follows: (1) proposing an efficient sequential inverse algorithm that can simultaneously determine several unknown time-dependent parameters; (2) exploring why the sequential function specification method can provide a stable but inaccurate estimation when tackling problems with larger measurement errors; and (3) discussing the sensitivity problem and analyzing what factors cause the growth in error sensitivity. Three examples are applied to demonstrate the performance of the proposed method, and the numerical results show that the accurate estimations can be obtained by alleviating the error sensitivity when the measurement error is considered.     

Parallel Newton methods for the nonlinear complementarity problem

In this paper, we discuss how the basic Newton method for solving the nonlinear complementarity problem can be implemented in a parallel computation environment. We propose some synchronized and asynchronous Newton methods and establish their convergence.This work was based on research supported by the National Science Foundation under grant ECS-8407240 and by a University Research and Development grant from Cray Research Inc. The research was initiated when the authors were with the University of Texas at Dallas.     

Controlling uncertain Lü system using backstepping design

In this paper, we discussed how to control Lü system with unknown parameters. Firstly we designed an observer to identify the unknown parameter of Lü system, then we used backstepping design method to control the system, and track any desired trajectory by the same way. At the same time we gave the numerical simulation for the results we had gained.     

POD a-posteriori error estimates for linear-quadratic optimal control problems

The main focus of this paper is on an a-posteriori analysis for the method of proper orthogonal decomposition (POD) applied to optimal control problems governed by parabolic and elliptic PDEs. Based on a perturbation method it is deduced how far the suboptimal control, computed on the basis of the POD model, is from the (unknown) exact one. Numerical examples illustrate the realization of the proposed approach for linear-quadratic problems governed by parabolic and elliptic partial differential equations.     

The unknown in decision making: What to do about it

The unknown or “other” that affects our lives is what we usually very much want to know about to cope with uncertainty. We often suspect that it affects us with partial and indefinite evidence that it exists but we only have uncertain feelings about it. Even when we do not know what it is we would like to allow for its influence in our explaining the outcome of a decision. One way to deal with the many factors of a decision is to include the unknown as one of them and then determine its priority of influence on the outcome by comparing it with other factors. We are able to do that to the extent that we are sure of what we know and of the residual that remains outside our understanding that may also have some effect on what we do. Confidence from good understanding and past success are what we need in order to judge the potential significance of what we do not know on the outcome. We can then perform sensitivity analysis to see how much effect unknown factors can have on the stability of the choice we make. Pairwise comparisons make it possible to tackle this idea explicitly and rather simply. This note illustrates how to prioritize and test the effect of the unknown alongside the known.     

Bounds on the number of steps of WDS required for checking the positivity of integral forms

Examples show that many integral forms can be efficiently verified to be positive through a special form of variable substitutions, i.e., weighted difference substitutions (WDS), but it was unknown how many steps of substitutions are needed, or furthermore, whether this method is adapted for checking positivity of all integral forms. In this paper, we give upper bounds of step numbers of WDS required in checking whether an integral form is positive or nonnegative, thus deducing that the positivity of integral forms can be completely verified through the WDS method.     

含参数线性方程组解的判别方法注记

当线性方程组中含有未知参数时,线性方程组解的情况往往需要进行讨论.本文给出了在非齐次线性方程组系数矩阵中含有未知参数且系数行列式等于零的情况下,判定对应参数值下方程组的解是无解还是有无穷多解的两个判定定理.和以前的方法比较,本文提出的讨论方法更直接.