多重线性的

The main theory results includes:(1) Using the properties of Hilbert transform, perfectly reconstruction and new type of lifting scheme, a new type of dual-tree binary coefficients complex wavelet with linear phase is achieved.(2) For linear systems that can be diagonalized by GFT and DST-II matrices, an efficient MGM method is proposed, convergence is proved.(3) We discuss the algebraic structure when Toeplitz matrix is transformed by multi-band wavelet,show that Toeplitz matrix is composed of generating function is transformed to a band and sparse matrix when wavelet applied to this matrix, based on the above results, an efficient solution of Toeplitz equations is obtained, and the computational complex is O,where N is the order of matrix.

理论成果主要包括：(1)对于对偶树二进制系数复数小波，利用Hilbert变换对性质、完全重构条件并结合新的提升格式构造研究了含参系数多进制小波构造方法，作为特例得到具有线性相位的对偶树二进制系数复数小波构造方法；(2)对于广义离散傅立叶变换与正弦变换对角化系统，提出了高效、快速的多重网格算法，理论上证明了算法的收敛性；(3)研究了Toeplitz矩阵在多进制小波变换下的代数结构，验证了多项式生成函数构成的Toeplitz系统在小波变换下的稀疏带宽性质，从而建立基于小波变换求解Toeplitz系统的快速求解方法，运算量级控制在O，其中N为系统的阶。

This paper is to mainly discuss the following problems which are diffieult to be solved in commodity freight volume forecast for a period of time: Variable choice and multiple coorelations; Application of the principles of combining quantitative analysis with qualitative analysis; Selection of the remarkable variables by using Multistep Regression Method; Establishment of reasonable freight volume forecast model.

本文针对总货运量预测中较难解决的问题——变量选择和多重共线性，运用定性和定量相结合原则利用逐步回归方法定量挑选显著性变量，建立理想的货运量预测模型。此外，本文在评介货流研究一般方法基础上，根据多地区货物交流目的，探讨了货流预测的方法，提出用Frator法预测货流。

Based on these and absorption of the foreign company performance results, the thesis combines the theory and the method to analyze the performance of electronic component manufacturing industry enterprises. Based on the analyzing system enterprise performance system in and abroad, the paper raises innovational achievement indexes and evolutional indexes of company performance as to the default of the electronic component manufacturing industry. The paper builds the hierachical clusteing method and bivariate correlations to choose the best evolutional indexes of company performance in order to avoid multicolinearity. The paper proposes the synthetic evaluating method principal component analysis with weighted variable by analytical process. Then, the paper applies the method to analyze the performance of 28 companies of electronic component manufacturing industry.

基于此，在消化吸收国外有关企业绩效评价方面的成果的基础上，针对企业绩效评价的缺陷，本文提出了企业绩效评价的成长性指标和价值性指标；针对改进的AHP与PCA，本文提出了改进的AHP加权主成分分析方法；由于指标之间可能存在着多重共线性，本文运用系统聚类分析与相关分析对企业绩效的成长性指标与价值性指标进行选取；运用改进的AHP加权主成分分析方法与因子分析对电子元器件制造业的28家上市公司进行实证分析，并验证本文提出的综合评价方法的优越性；本文运用灰色模糊评价方法对28家上市公司绩效的非财务指标进行评价；最后，给出相关的结论与建议。

The multiple corresponding analysis is suitable for correlation analysis of big sample categorical variable.

对于中医研究中常见的无序多分类或二分类变量，要同时研究它们之间的联系，用多重对应分析在结果的直观性和可解释性方面优于对数线性模型。

Then based on the analysis of relationship between variances and multi collinear, combined with systematic analysis of literatures and clinical experience, the logisti...

在变量的相关性及多重共线性分析并结合文献的系统分析及临床经验的基础上，应用SAS软件对有显著性意义的危险因素进行回归分析、聚类分析及主成分分析。

In chapter 7, chloride absorption or binding isotherm of concretes was investigated, linear chloride binding capacity and a non-linear coefficient were firstly proposed, based on a large amount of experimental data of concretes under the actions of single, double, or multiple factors in four typical bitterns.

第七章根据不同混凝土在4种典型盐湖的单一、双重和多重因素作用下的大量数据，研究了混凝土对氯离子的吸附/结合规律，提出了线性氯离子结合能力和非线性系数的新概念。

The researchmethodology system is as follows:The variables that can be measured and have significance to stock relative intrinsic value are industry characteristic, profitability,development potential,qulity of earing and asset, the expandable potential of capital stock,the ability to pay off the debt,asset manging ability and the scale of the capital stock;(2)When the MFPM is established, the original variables should be disposed beforehand;(3)As for the research objects.we should value synthesizationly the achievement of the listed,and take the No.

本章所建立的研究方法体系为：（1）上市公司个股基本面因素中可以量化，并且对股票相对内在价值有重要影响的主要有行业特性、盈利能力、成长性、盈利质量及资产质量、股本扩张潜力、偿债能力、资产运营效率和股本规模大小等方面；（2）在建立MFPM模型时，原始指标值是不能直接用于模型计算的，必须对数据进行预处理后才能用于模型中：（3）在确定研究对象时，需要对上市公司经营绩效进行综合评价，在剔除主要靠未来业绩预期定价的第Ⅳ类股后，把第Ⅰ、Ⅱ和Ⅲ殴作为模型研究对象；（4）在选择多元回归模型时，由于原始自变量由于存在错综复杂的关系，尽管对原始自变量进行过预处理，但自变量指标依然会存在一定的多重共线性，为此在建立MFPM模型时，须采用逐步回归模型；（5）在检验MFPM模型有效性时，须从统计意义和投资实战意义两个方面进行。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。