核实数据下误差在反映变量的非线性半参数模型的降维估计

考虑响应变量带有一般测量误差的非线性半参数模型.在核实数据的帮助下,利用半参数降维技术构造未知参数和非参数函数的估计.在一定条件下证明未知参数估计的渐近正态性和非参数函数估计的最优收敛速度.通过数值模拟说明所提估计方法在有限样本下的有效性.     

核实数据下非线性EV模型中经验似然降维推断

本文研究了响应变量有误差的非线性模型.应用半参数降维技术构造未知参数的被估计经验似然及调整的经验似然,证明了所提出的被估计的经验对数似然与其调整的经验对数似然分别渐近于独立卡方变量加权和的分布与标准卡方分布,所得结果可用来构造未知参数的置信域.     

非参数部分带有测量误差的部分线性变系数模型的经验似然推断

研究非参数部分带有测量误差的部分线性变系数模型,构造了模型中未知参数的局部纠偏经验对数似然比统计量,在适当条件下,证明了所提出的统计量具有渐近x2分布,由此结果可以用来构造未知参数的置信域.并且还构造了未知参数的最大经验似然估计及系数函数的估计,证明了它们的渐近性质.最后通过数值模拟研究了所提估计方法在有限样本下的实际...     

缺失数据下非线性半参数EV模型的估计

考虑解释变量带有测量误差且响应变量随机缺失情形下的非线性半参数EV模型. 利用核实数据,构造了未知参数和非参数函数的两种估计.证明了未知参数估计的渐近正态性,给出了非参数函数估计的最优收敛速度.     

带Gauss型误差的GMANOVA-MANOVA模型中未知参数的精确置信域

对于带Gauss型误差的GMANOVA-MANOVA模型,在均匀协方差结构下,求出了其中未知参数的极大似然估计及其均值和方差,并依据极大似然估计构造了未知参数的精确置信域.     

纵向数据下部分非线性模型的广义经验似然推断

本文研究了纵向数据下部分非线性模型中未知参数的置信域的构造.利用经验似然方法,构造了非线性函数中未知参数的广义对数经验似然比统计量,证明了其渐近于卡方分布.同时,得到了未知参数的最大经验似然估计,并证明了其渐近正态性.     

具有Rao简单结构的多元t-模型的参数区间估计

在统计推断中,模型中未知参数的区间估计是一个重要的研究内容.本文通过构造枢轴量研究了具有Rao简单结构多元￡-模型的参数区间估计,利用条件分布技巧,得到枢轴量的分布,从而获得模型中单个未知参数的置信区域以及参数的联合置信区域.     

均匀分布参数的无偏估计及其分布

讨论了均匀分布未知参数无偏估计量的分布密度,利用无偏估计量构造出一些新的样本函数,并且利用给出的样本函数推导出了未知参数的置信区间.所得到结果改善了现有的估计,易于计算.     

重尾非线性自回归模型自加权M-估计的渐近分布

考虑非线性自回归模型xt=f(xt-1,…,xt-p,θ)+∈t,其中θ为q维未知参数,{∈t}为随机误差.在允许误差方差无穷的重尾条件下,构造θ的自加权M-估计,并证明了该估计的渐近正态性.最后通过数值模拟,在随机误差服从某些重尾分布的条件下,说明自加权M-估计比最小二乘和L1估计更有效.     

非线性自回归模型的分位数推断

考虑非线性自回归模型x_t=f (x_(t-1),···, x_(t-p),θ)+ε_t,其中f (·)为R~p上的实值可测函数,θ为q维未知参数,{ε_t}为随机误差.构造了θ的分位数估计,并证明了该估计的渐近正态性,最后通过数值模拟,说明了该估计的稳健和有效性.     

Parameter estimation by Hellinger type distance for multivariate distributions based upon probability generating functions

Maximum likelihood (ML) estimation is a popular method for parameter estimation when modeling discrete or count observations but unfortunately it may be sensitive to outliers. Alternative robust methods like minimum Hellinger distance (MHD) have been proposed for estimation. However, in the multivariate case, the MHD method leads to computer intensive estimation especially when the joint probability density function is complicated. In this paper, a Hellinger type distance measure based on the probability generating function is proposed as a tool for quick and robust parameter estimation. The proposed method yields consistent estimators, performs well for simulated and real data, and can be computationally much faster than ML or MHD estimation.     

Parameter estimation of delay differential equations: An integration-free LS-SVM approach

This paper introduces an estimation method based on Least Squares Support Vector Machines (LS-SVMs) for approximating time-varying as well as constant parameters in deterministic parameter-affine delay differential equations (DDEs). The proposed method reduces the parameter estimation problem to an algebraic optimization problem. Thus, as opposed to conventional approaches, it avoids iterative simulation of the given dynamical system and therefore a significant speedup can be achieved in the parameter estimation procedure. The solution obtained by the proposed approach can be further utilized for initialization of the conventional nonconvex optimization methods for parameter estimation of DDEs. Approximate LS-SVM based models for the state and its derivative are first estimated from the observed data. These estimates are then used for estimation of the unknown parameters of the model. Numerical results are presented and discussed for demonstrating the applicability of the proposed method.     

State estimation in batch processes using a nonlinear observer

This paper deals with the problem of nonlinear states estimation in batch chemical processes. It presents a reduced-order nonlinear observer approach to perform the estimation. The proposed method allows adjustment of the speed of convergence towards zero of the estimation error. The stability properties of the model-based observer are analytically treated in order to show the conditions under which exponential convergence can be achieved. In addition, the performance of the proposed observer is evaluated on batch processes.     

Adjustable dimension descriptor observer based fault estimation for switched nonlinear systems with partially unknown nonlinear dynamics

This paper focuses on the fault estimation problem for switched systems with partially unknown nonlinear dynamics, actuator and sensor faults, simultaneously. The fault estimation observers are constructed, in which the observer dimension is not fixed and can be selected in a certain range. Both the disturbance decoupling and disturbance attenuation are considered, where the unknown nonlinear dynamics can be decoupled and the effect of modeling error and measurement disturbance is attenuated. Based on the average dwell time and the piecewise Lyapunov function, the observer parameter matrices can be calculated by solving LMIs and matrix equations. Finally, two examples are listed to verify the proposed fault estimation approach.     

Robust state estimation and fault diagnosis for uncertain hybrid nonlinear systems

Robust state estimation and fault diagnosis are challenging problems in the research of hybrid systems. In this paper, a novel robust hybrid observer is proposed for a class of uncertain hybrid nonlinear systems with unknown mode transition functions, model uncertainties and unknown disturbances. The observer consists of a mode observer for discrete mode estimation and a continuous observer for continuous state estimation. It is shown that the mode can be identified correctly and the continuous state estimation error is exponentially uniformly bounded. Robustness to unknown transition functions, model uncertainties and disturbances can be guaranteed by disturbance decoupling and selecting proper thresholds. The transition detectability and mode identifiability conditions are rigorously analyzed. Based on the robust hybrid observer, a robust fault diagnosis scheme is presented for faults modeled as discrete modes with unknown transition functions, and the analytical properties are investigated. Simulations of a hybrid three-tank system demonstrate that the proposed approach is effective.     

Correct algebras over estimation algorithms in the set of regular recognition problems with nonoverlapping classes

Algebras over estimation algorithms in the set of regular problems with nonoverlapping classes are considered. A correctness criterion for the arbitrary degree algebraic closure of the model of estimation algorithms in the classification problems of this type is proposed; this criterion can be efficiently verified. An estimate of the minimal degree of the algebraic closure that is sufficient for constructing a correct classifier in an arbitrary regular problem with nonoverlapping classes is found.     

随机缺失数据下样本分位数估计

分位数的估计在生物医学、社会经济调查等领域有着广泛的应用,然而在实际问题的研究中,往往由于各种人为或不可控因素造成数据收集不完全.本文在随机缺失(MAR)假设条件下,利用非参数核补法和局部多重插补法给出了响应变量缺失时样本分位数的估计,并利用经验过程等理论证明了由这两种方法得到的分位数估计的大样本性质,同时,使用重抽样方法给出了估计的渐近方差的估计,模拟结果验证了这两种方法的有效性.文章所提两种方法的优点在于:首先,所提出的缺失修正方法不需要对缺失概率的模型做任何假设;其次,方法亦适用于其他有关参数不可微的估计目标函数;最后,方法很容易地推广到一般M估计的情况,并可以对多个分位数同时进行估计.     

系数函数光滑度互不相同的变系数模型的一步估计法

该文提出了一种一步估计方法用以估计变系数模型中具有互不相同光滑度的未知函数, 所有未知函数和它们的导数的估计量由 一次极小化得到. 给出了估计量的渐近性质, 包括渐近偏差、方差和渐近分布, 一步估计量被证明达到了最优收敛速度.     

Approximate Bayesian Approach to Non-Gaussian Estimation in a Linear Model with Dependent State and Noise Vectors

This paper extends the results of Masreliez [8] on the design of non-Gaussian estimators for a more general class of the parameter estimation problem when the system state and the observation noise may be dependent and non-Gaussian simultaneously. It is shown that the proposed non-Gaussian algorithms can approximate with high precision the minimum mean square estimator. Application of the approach to the design of different optimal (and stable) estimation algorithms is illustrated. The efficiency of the proposed algorithms is tested in some simulation experiments. Accepted 5 September 2000. Online publication 26 February 2001.     

Algebraic switching time identification for a class of linear hybrid systems

In this paper, a method for the finite time estimation of the switching times in linear switched systems is proposed. The approach is based on algebraic tools and distribution theory. Switching time estimates are given by explicit algebraic formulae that can be implemented in a straightforward manner using standard tools from computational mathematics. Simulations illustrate the proposed techniques.