Parameter cascades and profiling in functional data analysis

A data smoothing method is described where the roughness penalty depends on a parameter that must be estimated from the data. Three levels of parameters are involved in this situation: Local parameters are the coefficients of the basis function expansion defining the smooth, global parameters define low-dimensional trend and the roughness penalty, and a complexity parameter controls the amount of roughness in the smooth. By defining local parameters as regularized functions of global parameters, and global parameters in turn as functions of complexity parameter, we define a parameter cascade, and show that the accompanying multi-criterion optimization problem leads to good estimates of all levels of parameters and their precisions. The approach is illustrated with real and simulated data, and this application is a prototype for a wide range of problems involving nuisance or local parameters.     

Rapid parameter identification of linear time-delay system from noisy frequency domain data

The technique to identify the system parameters thereof has attracted extensive research interest, since knowing the parameters would enable effective system control strategy and accurate response prediction. In this paper, a novel approach is developed to identify the parameters of the linear time-delay differential system by analyzing the complex system response in the frequency domain. Firstly, the complex frequency response of the time-delay system is expressed as a function of physical parameters and time-delay parameters, forming a typical optimization problem. Subsequently, the sensitivities with respect to the unknown parameters are derived. A novel sensitivity-based algorithm is adopted in the identification procedure. Trust-region constraint is implemented and hence tackled by Tikhonov regularization, which effectively enhances the efficiency of the algorithm. The feasibility and robustness of the identification procedure are evaluated by identifying the parameters of two numerical time-delay systems and an experimental case.     

Detecting changes in the ar parameters of a nonstationary arma process

We present a method for detecting changes in the AR parameters of an ARMA process with arbitrarily time varying MA parameters. Assuming that a collection of observations and a set of nominal time invariant AR parameters are given, we test if the observations are generated by the nominal AR parameters or by a different set of time invariant AR parameters. The detection method is derived by using a local asymptotic approach and it is based on an estimation procedure which was shown to be consistent under nonstationarities.     

Analysis of correlated binary data under partially linear single-index logistic models

Clustered data arise commonly in practice and it is often of interest to estimate the mean response parameters as well as the association parameters. However, most research has been directed to address the mean response parameters with the association parameters relegated to a nuisance role. There is relatively little work concerning both the marginal and association structures, especially in the semiparametric framework. In this paper, our interest centers on the inference of both the marginal and association parameters. We develop a semiparametric method for clustered binary data and establish the theoretical results. The proposed methodology is investigated through various numerical studies.     

Effects of system parameters on the optimal policy structure in a class of queueing control problems

This paper studies a class of queueing control problems involving commonly used control mechanisms such as admission control and pricing. It is well established that in a number of these problems, there is an optimal policy that can be described by a few parameters. From a design point of view, it is useful to understand how such an optimal policy varies with changes in system parameters. We present a general framework to investigate the policy implications of the changes in system parameters by using event-based dynamic programming. In this framework, the control model is represented by a number of common operators, and the effect of system parameters on the structured optimal policy is analyzed for each individual operator. Whenever a queueing control problem can be modeled by these operators, the effects of system parameters on the optimal policy follow from this analysis.      

Inverse analysis of an internal reforming solid oxide fuel cell system using simplex search method

An inverse problem is solved for estimating fuel cell operating parameters such as current density, pressure and fuel flow rate (FFR) separately and then simultaneously two parameters in an internal reforming solid oxide fuel cell (IRSOFC). Initially, a mathematical model for the forward problem is developed to simulate the IRSOFC steady state operation and its performance in terms of power output and then an inverse problem is solved for recovering the above parameters using a simplex search minimization algorithm. The objective function (IRSOFC power) and the estimation accuracy are studied for the effects of initial guess values of the operating parameters and the number of iterations required for retrieval of these parameters. The objective function is represented by the sum of square of the error between a given IRSOFC power and the power evaluated based on some arbitrary guessed values of the unknowns which is then regularized in an iterative manner for solution of the inverse fuel cell problem. The study reveals that a multiple combinations of parameters (current density, operating pressure and FFR) exist which provides guidelines for selecting feasible combinations of these parameters required for meeting a given power requirement. The results show relatively good agreement between the inverse and exact solutions.     

基于贝叶斯平滑转移模型的汇率非线性协整关系研究

针对平滑转移模型参数估计不确定性导致的协整检验方法相对复杂问题,提出基于平滑转移模型的贝叶斯非线性协整分析。通过模型的统计结构分析,选择参数先验分布,结合参数的后验条件分布特征设计Metropolis-Hasting-Gibbs混合抽样方案,据此估计平滑转移模型的参数,并对回归残差进行贝叶斯单位根检验,解决参数估计过程中遇到的参数估计不确定性及协整检验复杂的问题;利用人民币对美元汇率与中美两国的利率数据进行实证分析。研究结果表明:MH-Gibbs抽样方案能够有效估计平滑转移模型的参数,中美汇率波动和利差之间存在平滑转移协整关系。     

Damage identification in a concrete dam by fitting measured modal parameters

We propose a method, based on an inverse problem, to obtain numerically the material parameters that characterize the elasticity tensor of a body with linear elastic behavior, using accurate measurements of the first modal parameters, namely the natural frequencies and the modes of vibration (the eigenfrequencies and the eigenvectors). Appropriate functionals are defined, whose minimum points correspond to the unknown material parameters. To obtain these minimum points a highly nonlinear parametric optimization problem is solved. Its resolution involves specific mathematical tools like the derivative of the eigenvalues and eigenvectors with respect to the material parameters, the adjoint method, and gradient methods for the minimization of the functional. An application is presented, which considers a cracked dam in which is assumed the presence of transversely isotropic material in the cracked zone. The material parameters of the transversely isotropic material are obtained by minimizing the distance between the modal parameters (eigenfrequencies and eigenvectors) of a numerical model of the dam and the observed modal parameters physically measured in the dam. The algorithm is implemented in a C++ home made code with the aid of open-source libraries for scientific computation.     

Stability sets of multiparameter Hamiltonian systems

A real linear Hamiltonian system with constant coefficients that depend on several real parameters is considered. A method is proposed for calculating the sets of all values of the parameters for which the stationary solution of this system is stable for fixed values of the parameters (that is, the stability sets). The application of the method is demonstrated for a gyroscopic problem described by a Hamiltonian system with four degrees of freedom and three parameters. Computer algebra, in particular, a Gröbner basis and a Power Geometry are used. It is shown that the four-parameter generalization of this problem does not contain fundamentally new difficulties.     

Semiparametric marginal and association regression methods for clustered binary data

Clustered data arise commonly in practice and it is often of interest to estimate the mean response parameters as well as the association parameters. However, most research has been directed to inference about the mean response parameters with the association parameters relegated to a nuisance role. There is little work concerning both the marginal and association structures, especially in the semiparametric framework. In this paper, our interest centers on inference on the association parameters in addition to the mean parameters. We develop semiparametric methods for both complete and incomplete clustered binary data and establish the theoretical results. The proposed methodology is illustrated through numerical studies.     

Gradient-based Regularization Parameter Selection for Problems With Nonsmooth Penalty Functions

In high-dimensional and/or nonparametric regression problems , regularization (or penalization) is used to control model complexity and induce desired structure. Each penalty has a weight parameter that indicates how strongly the structure corresponding to that penalty should be enforced. Typically, the parameters are chosen to minimize the error on a separate validation set using a simple grid search or a gradient-free optimization method. It is more efficient to tune parameters if the gradient can be determined, but this is often difficult for problems with nonsmooth penalty functions. Here, we show that for many penalized regression problems, the validation loss is actually smooth almost-everywhere with respect to the penalty parameters. We can, therefore, apply a modified gradient descent algorithm to tune parameters. Through simulation studies on example regression problems, we find that increasing the number of penalty parameters and tuning them using our method can decrease the generalization error.     

Dynamic inversion method for the material parameters of a high arch dam and its foundation

The overall mechanical behavior of the structure of an arch dam is comprehensively reflected by the vibration modal information included in measured vibration response. Hence, the results obtained from inverting material parameters based on measured vibration data are often superior to those based on static monitoring data. In this study, a dynamic inversion method for the material parameters of a high arch dam and its foundation is proposed on the basis of the measured vibration response. First, an arch dam prototype test is conducted to obtain the measured dynamic displacement response as input. Then, a stochastic subspace identification method based on singular entropy is formulated to determine the modal parameters. Second, a dynamic elastic modulus (DEM) with a great influence on the modal parameters is selected as the material parameter to be inverted. Then, a response surface model (RSM), which reflects the nonlinear relationship between the material and modal parameters of each zone, is constructed. Latin hypercube sampling is used to generate the sample library of the DEM. The RSM is fitted by modal parameters calculated on the basis of the arch dam finite element model (FEM) and is applied to replace the FEM. Finally, the optimization mathematical model of the inversion of the DEM is established. Then, the objective function is optimized through a genetic algorithm, and the optimal combination of the DEM in each zone is inverted. The modal parameters of the arch dam calculated by inversion results are consistent with those measured by variation law and values. Therefore, the inversion results are reasonable and reliable. This method provides a new idea for determining the material parameters of a high arch dam and its foundation during the operation period.     

基于四参数区间数的多准则决策方法

针对准则权重和准则值均为四参数区间数的多准则决策问题,提出了一种基于四参数区间集结算子的多准则决策方法。首先,定义了四参数区间数以及加权的连续四参数区间数有序加权平均 (WCFP-OWA) 算子和加权的连续四参数区间数有序加权几何平均 (WCFP-OWG) 算子,并利用连续四参数区间数有序加权平均 (CFP-OWA) 算子对四参数区间数准则权重进行处理,然后利用WCFP-OWA算子和WCFP-OWG算子对四参数区间数准则值进行集成得到综合值以确定方案排序。最后将该方法应用于勘探公司的选择,验证了该方法的合理性和有效性。     

An implicit least squares algorithm for nonlinear rational model parameter estimation

In this study a new insight into least squares regression is identified and immediately applied to estimating the parameters of nonlinear rational models. From the beginning the ordinary explicit expression for linear in the parameters model is expanded into an implicit expression. Then a generic algorithm in terms of least squares error is developed for the model parameter estimation. It has been proved that a nonlinear rational model can be expressed as an implicit linear in the parameters model, therefore, the developed algorithm can be comfortably revised for estimating the parameters of the rational models. The major advancement of the generic algorithm is its conciseness and efficiency in dealing with the parameter estimation problems associated with nonlinear in the parameters models. Further, the algorithm can be used to deal with those regression terms which are subject to noise. The algorithm is reduced to an ordinary least square algorithm in the case of linear or linear in the parameters models. Three simulated examples plus a realistic case study are used to test and illustrate the performance of the algorithm.     

Estimation in the presence of many nuisance parameters: Composite likelihood and plug-in likelihood

We consider the incidental parameters problem in this paper, i.e. the estimation for a small number of parameters of interest in the presence of a large number of nuisance parameters. By assuming that the observations are taken from a multiple strictly stationary process, the two estimation methods, namely the maximum composite quasi-likelihood estimation (MCQLE) and the maximum plug-in quasi-likelihood estimation (MPQLE) are considered. For the MCQLE, we profile out nuisance parameters based on lower-dimensional marginal likelihoods, while the MPQLE is based on some initial estimators for nuisance parameters. The asymptotic normality for both the MCQLE and the MPQLE is established under the assumption that the number of nuisance parameters and the number of observations go to infinity together, and both the estimators for the parameters of interest enjoy the standard root-n$n$ convergence rate. Simulation with a spatial–temporal model illustrates the finite sample properties of the two estimation methods.     

Optimal finite difference schemes for a wave equation

In this paper, a solution to a two-dimensional wave equation using the Laguerre transform is considered. Optimal parameters of finite difference schemes for this equation are obtained. Numerical values of these optimal parameters are specified. Second-order finite difference schemes with the optimal parameters provide an accuracy of solving the equations close to that provided by a fourth-order scheme. It is shown that using the Laguerre decomposition can reduce the number of optimal parameters in comparison with using the Fourier decomposition. This simplifies the finite difference schemes and decreases the number of calculations, that is, makes the algorithm more efficient.     

复共线联立方程模型参数的修正间接岭估计

在结构方程恰好被识别时,研究了外生变量设计矩阵X复共线时联立方程模型的参数估计问题,提出了参数的一种修正间接岭估计方法,并证明了这种参数估计的良好统计性质,最后给出了在修正间接岭估计均方误差最小意义下岭参数的一种选择方法.     

Schematic 4-designs

We study 4-designs with three intersection numbers. By the Cameron-Delsarte theorem, the blocks form a symmetric three-class association scheme. This imposes strong restrictions on the parameters of such designs. We calculate the eigenvalues of the association scheme from the design parameters and determine all admissible parameters with at most 1000 points. An infinite family of admissible parameters is discovered. Designs with small admissible parameters exist and are related to the quadratic residue codes.     

Fuzzy-Based Analysis of a Hill-Type Muscle Model

Among the most challenging problems in present day mechanics is the realistic simulation of the human kinetic behavior which requires correct modeling of the muscles – the human actuators. We are, therefore, implementing and validating a 1-D massless Hill-type muscle model in the program Neweul-M². As a part of this effort, various anatomical and physiological parameters need to be chosen prior to the simulation. These parameters are always subject to natural variability or scatter. Therefore, it is important to consider those parameters as uncertain fuzzy parameters and then determine the sensitivity measures of the parameters involved in various simulation scenarios using the program Famous. The applied sensitivity analysis quantifies the influence of individual parameters or subsets of parameters on the output quantity of interest. Thus, this analysis identifies those parameters with large influence on the simulation response of muscles which, therefore, needs to be chosen very carefully. (© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)