回归估计

In this paper, the function of regression matrix is estimated in a growth curve model, and the necessary and sufficient conditions that a linear estimator of the function is admissible among the class of all lin ear estimators under matrix loss are obtained.

摘 要：对增长曲线模型中回归矩阵的函数的线性估计进行了研究。在矩阵损失下，作者得到了线性估计在一切线性估计类中可容许的充要条件。

The common method, that all strong-correlation terms of the model are eliminated, can bring the loss in the engineering application, so the new method is proposed that the identified model reserves some correlation. The augmented matrix A is constructed by the outputΔW and the matrix S. The"determinating order based on ratio of determinant"is brought out to screen the strong-correlation terms in the structure identification. The latent root estimation is improved in screening the eigenvalues and eigenvectors. Thus the estimation precision is improved greatly.The consistence check of guidance instrument error coefficients of flight test and ground test is the purpose of flight experiment. The causes of inconsistency of the two models are analyzed. The hypothesis test of linear regression model based on F statistics is proposed to check the consistence.Finally, the instability of error coefficients is probably caused by the change of the flight environments, therefore, the relation between the error coefficients and flight environment is analyzed. The approach is presented to identify SINS guidance instrument error models and compensate the error in the segmented sections corresponding to the change of vertical acceleration of aircraft.

在结构辨识中，常用的方法由于将模型中的强相关项全部剔除而给工程应用带来损失，因此，本文提出了新的有益思想，即在保留一定相关性的基础上进行辨识：将输出向量ΔW与环境函数矩阵S构成增广矩阵A，然后采用"比定阶行列式"来剔除相关向量的方法，这样既可以尽可能多地保留了对落点影响大的强相关参数，又可以对落点影响小的强相关参数给予剔除；在参数估计中，改进了特征根估计中特征根和特征向量的筛选方法，提出"近零"准则，从而大大提高了参数估计的精度；再者，鉴于天地模型"一致性"检验是飞行试验和SINS制导工具误差系数分离的主要目的，因此，本文又深入分析了造成天地模型不一致的原因，提出了采用基于F统计的线性回归模型假设检验方法来进行捷联制导工具误差模型的天地"一致性"检验；最后，鉴于飞行环境剧烈变化可能会对惯性仪表误差系数稳定性带来一定的影响，因此本文深入地分析了SINS制导工具误差系数与外界环境的关系，提出了基于过载变化大小的分段辨识和分段实时补偿的算法。

There are two findings summarized from previous studies:(1) the least squares estimator is a common choice of researchers, but under an unequal probability design, the estimator is biased,(2) the probability weighted estimator is consistent but may have a large variance.

过去的研究发现：(1)在不等机率抽样时，一般常用的最小平方法所得到的回归参数估计式并非不偏估计式；(2)机率加权最小平方法可以改善不偏性，然却增加估计的变异程度。

As to seemingly unrelated regression system,a new biased contracting estimator of the parameters is put forward,which is the combination of generalized ridge covariance-improved estimator and Stein estimator,and the good features of this estimator in mean square error is discussed.

对于一类相依回归系统，结合广义岭型协方差改进估计与Ste in估计，提出了一种新的有偏压缩估计，，并讨论了该估计在均方误差下的优良性质

In this paper we discuss the influence problem of Ridge Estimator in general linear regression model, study the influence analysis on Ridge Estimator with covariance matrix disturbance and data deletion, define Cook distance of Rideg Estimator based on biased estimator.

讨论一般线性回归模型岭估计的影响+分析问题，研究了协方差阵扰动和数据删除对岭估计的影响，给出了岭估计基于有偏估计的Cook距离。

The Minimax admissibility of linear estimates with respect to restricted multivariate regression coefficient under matrix loss function is considered.

在矩阵损失下给出了带约束的多元回归系数线性估计在线性估计类中是 Minimax可容许估计的充要条件。

In this paper the spline regressive analysis is used for the information processing of multi-gases sensor signals with the electronic nose system. The authors have established a spline regressive model of the freshness degree of beef based on the beef sample, and have done regressive analysis to the model, which has given the forecast value and the estimative interval of the degree of the freshness of the beef.

将样条回归分析方法用於电子鼻系统中多气体感测器信号的资讯处理，针对牛肉样本建立了牛肉新鲜度样条回归模型，并对该模型进行了回归分析给出了牛肉新鲜预报值和估计区间。

Finally, to explain the advantage and shortcoming of nonparametric analysis and parametric analysis, and introduce some main shapes of nonpametric regression model: Fourier series, kernel estimators, and spline smoothing. In the meantime, we give the approximate theory and approximate distribution theory of these methods.

阐述了非参数回归分析和参数回归分析的优缺点，介绍了非参数回归模型的几种主要形式：富氏级数，核估计量和样条平滑，给出了这些方法的逼近理论和逼近分布情况。

Based on the characteristics that the generalized regression neural network converges to the underlying regression surface and can be used for any regression problem, this dissertation exploits GRNN to investigate large-scale IP traffic matrix estimation problem.

2基于广义回归神经网络(Generalized Regression NeuralNetwork，GRNN)收敛于基本的回归面和能用于处理任何回归问题的特性，本文利用GRNN来研究大尺度IP流量矩阵估计问题。

The first one focuses upon the fundamental theory and methods of system identification, including the basic concepts of system description and identification,review of matrix theory, stochastic process and pulse response of linear systems; the first principle models for lumped and distributed parameter systems; modeling of bioreactions; auto regressive moving average model; weighted least square and regressive weighted least square for parameter estimation of ARMA model; Runge-Kutta and Simplex methods; an example of BPNN application; as well as Kalman Filter.

第一部分包括知识驱动建模、数据驱动建模、参数估计和状态估计的基础理论和方法。具体为：(1)系统辨识和系统描述的基本概念，包括矩阵论、随机过程、线性系统的脉冲响应函数等基础知识回顾；(2)基于动态物料和能量衡算的集总参数、分布参数对象的知识驱动建模方法；(3)生物反应系统的建模方法；(4)自回归移动平均模型；(5) ARMAX模型参数的加权最小二乘参数估计一次性算法和递推算法，相应的统计性质；(6)龙格库塔法和单纯形法；BP神经网络应用实例；(7)卡尔曼滤波器，及其与加权最小二乘参数估计递推算法的对比。

On the other hand, the more important thing is because the urban housing is a kind of heterogeneity products.

另一方面，更重要的是由于城市住房是一种异质性产品。

Climate histogram is the fall that collects place measure calm value, cent serves as cross axle for a few equal interval, the area that the frequency that the value appears according to place is accumulated and becomes will be determined inside each interval, discharge the graph that rise with post, also be called histogram.

气候直方图是将所收集的降水量测定值，分为几个相等的区间作为横轴，并将各区间内所测定值依所出现的次数累积而成的面积，用柱子排起来的图形，也叫做柱状图。