回归的

Based on the autoregression equation for the standard variation of the residual errors obtained by ordinary regression analysis, the least square estimates and the maximum likelihood estimates for the regression, the autoregression and the moving average coefficients are derived.

提出一种异方差回归－时序模型，通过建立回归分析残差的标准差自回归方程，给出回归系数、自回归系数和滑动平均系数的最小二乘估计和极大似然估计。

The contents from the third chapter to the fifth chapter are the kernels which are a series of applications of generalized regression model and generalized regression estimator. At first, this paper constructs ratio model, linear regression model, post-stratified regression model and nonparametric regression model through different regressive relationship between auxiliary variable and study variable.

首先是依据辅助变量与研究变量之间回归关系的不同，分别建立比率模型、线性回归模型、事后分层回归模型和非参数回归模型，然后再利用第二章中推导出的广义回归估计理论，对各种回归模型辅助下(来源：A4bBC12论文网www.abclunwen.com)的估计方法进行了系统性研究，从而构成了模型辅助估计方法的整个研究体系。

Combining the experimental results and theoretical analysis, the following work has been accomplished.(1) Reasons of the decrease of interelectrode capacitance and the increase of discharge gap caused by the powder suspended in dielectric fluid are explained. And the mechanism that surface quality of workpiece improved by powder-mixed EDM is analyzed from this viewpoint.(2) The variation regularity of the percentage of silicon on the machined surface of workpiece has been summarized according to the experiment results when silicon powder was added into the working fluid. The entropy is introduced to explain the reasons for this variation regularity by thermodynamics.(3) The dimensional accuracy is greatly influenced by the overcut resulting from discharge gap and electrode wear in practice application. Empirical formulas for the overcut in EDM and powder-mixed EDM are established based on the experimental results using the ternary linear regressive method in this paper.(4) The influence of main factors on the machined surface characteristics has been evaluated based on the orthogonal experimental design method.

通过一系列普通电火花加工与混粉电火花加工的对比实验，运用工艺实验和理论分析相结合的方法，重点做了以下工作：(1)探讨了粉末颗粒导致极间电容减小、放电间隙增大的原因，并从这个角度解释了混粉电火花加工可以改善工件表面质量的机理；(2)以混合硅粉的电火花加工为例，实验得出峰值电流和脉冲宽度对加工表面层成分中硅元素含量的影响规律，并引入熵的概念从热力学的角度对硅元素含量变化的原因寻找了理论依据；(3)在实际加工中，由于粉末的加入使放电间隙变大，这使得混粉电火花加工的尺寸精度难以控制，本文利用三元线性回归的方法，在实验数据的基础上，建立了两种电火花加工过切量的计算公式；(4)将正交试验法应用于混粉电火花加工实验研究，并通过对实验指标极差的直观分析，得出各电参数对加工指标影响的敏感程度。

Chapter 2: Expatiate on the basic theory of the GRIN micro-sphere-lens, especially on the classical model—MAXWELL fisheye lens, which demonstrates their excellent optical properties.Chapter 3: Simply introduced the concept of retro-reflection, retro-reflection materials, the fabrication of traditional glass sphere and the possibility of the application of GRIN polymer micro-sphere on retro-reflection.

第三章：介绍了回归反射的概念、回归反射材料的现状；研究了用麦克斯韦鱼眼微球制作回归反射材料的可能，并与传统的均匀折射率玻璃微球回归反射材料的性能作了比较，得出将麦克斯韦鱼眼微球用作回归反射材料不仅可以大幅度降低原先对材料折射率的过高要求，还可明显提高回归反射的效率的结论。

For a seemingly unrelated regression system consisting of m equations, we propose a new type of estimator called pre test ridge regression estimator by combining the idea of preliminary test and ridge regression methodology and establish some statistical properties of PTRRE.

对于由 m个相依线性回归方程组成的线性回归系统，本文作者基于岭回归估计和预检验估计，提出了回归参数的一种新型估计，即预检验岭回归估计。并讨论了它的统计性质。

Based on the phase space model of the zeroth and first order approximation, a regression-autoregression model and a regression-local linear regression model are proposed, which provided new method for dam safety monitoring.

在此基础上，提出了逐步回归—自回归、逐步回归—局域线性回归等大坝观测序列相空间模型，为大坝安全监控模型的建立与预测提供了新的思路与方法。

On the basis of the theoretic analysis, this paper regards to the swatch of the listed corporate of our country in Shanghai Stock Exchange from 2002 to 2004, and sets up the model by linear regression and non-linear regression to analysis the listed company performance index sign and the financing structure, and carries through supposing examination, draws the conclusions:(1) The creditor's rights takes on the characteristic of a soft control and infirmness in the Chinese listed company governance, without bringing into play an obligatory function, the liability financing exists the best financing zone.

在理论分析的基础上，本文以2002―2004年度我国沪市上市公司为样本展开实证研究，采用线性回归和非线性回归、一元回归和多元回归等多种方法建立模型，进行上市公司绩效指标与融资结构的回归分析，并进行假设检验，得出如下结论：(1)债权在中国上市公司治理中表现出软约束和弱化的特征，没有发挥出应有的作用，债权融资存在最优融资区间。

Unitary linear regression is done between w~e、 w~ρ、 e~ρ、w_L~w_P. For α_v 、 C_ c between w、ρ、 e、 w_P in eastern China, unitary linear regression is done to silty clay as well as multivariant regression with coefficient iteration method and least-squares procedure are done to all kinds of soil, and the two results are compared. Unitary linear regression and multivariant regression are done to c_q、φ、 c_d、φ_d between e and I_P. Unitary linear regression and multivariant linear regression are done to napierian logarithm of q_u between e 、 I_L、 w_P and w_L, then the equations of regression are checked with other parameters.

对物理指标含水量～孔隙比、含水量～湿密度、孔隙比～湿密度、液限～塑限作了一元线性回归；对华东地区粉质粘土的压缩系数、压缩指数和含水量、湿密度、孔隙比及塑性指数作了一元线性回归；对华东地区所有土类的压缩系数、压缩指数和物理性质参数用系数迭代法和最小二乘法作了多元线性和非线性回归，并对两种方法得到的回归结果作了比较；作了快剪强度指标和固结快剪强度指标关于孔隙比和塑性指数的一元线性回归和多元线性及非线性回归；对无侧限抗压强度的自然对数值和孔隙比、液性指数、液限及塑限作了一元和多元线性回归，并用其它工程的参数对回归方程进行了检验。

The article present the conception of regression testing and the step of executing this testing. Introduce how to maintenance the test case library which used in regression testing ,and provide the method of ensure regression testing's validity. Finally, it gives some problem must be careful in the period of regression

本文描述了软件回归测试的概念和进行回归测试的基本步骤，介绍了可用于回归测试的测试用例库的维护方法，给出了几种可以可保证回归测试效率和有效性的回归测试策略，总结了回归测试时应该注意的一些实际问题。

We introduce an accurate online-learning support vector machine algorithm, including incremental and decremental algorithm, which efficiently updates a trained regression function whenever a sample is added to or removed from the training set.

本文充分利用支撑向量机解的稀疏性并结合KKT条件，当向训练样本集中添加新样本时，调用&递增&算法修改原有回归函数来得到新回归函数；当从训练样本集中剔除旧样本时，调用&递减&算法修改原有回归函数得到新回归函数，即在线学习方法。

This one mode pays close attention to network credence foundation of the businessman very much.

这一模式非常关注商人的网络信用基础。

Cell morphology of bacterial ghost of Pasteurella multocida was observed by scanning electron microscopy and inactivation ratio was estimated by CFU analysi.

扫描电镜观察多杀性巴氏杆菌细菌幽灵和菌落形成单位评价遗传灭活率。