A semiparametric model for compositional data analysis in presence of covariates on the simplex

Compositional data occur as natural realizations of multivariate observations comprising element proportions of some whole quantity. Such observations predominate in disciplines like geology, biology, ecology, economics and chemistry. Due to unit sum constraint on compositional data, specialized statistical methods are required for analyzing these data. Dirichlet distributions were originally used to study compositional data even though this family of distribution is not appropriate (see Aitchison, 1986) because of their extreme independence properties. Aitchison (1982) endeavored to provide a viable alternative to existing methods by employing Logistic Normal distribution to analyze such constrained data. However this family does not include the Dirichlet class and is therefore unable to address the issue of extreme independence. In this paper generalized Liouville family is investigated to model compositional data which includes covariates. This class permits distributions that admit negative or mixed correlation and also contains non-Dirichlet distributions with non-positive correlation and overcomes deficits in the Dirichlet class. Semiparametric Bayesian methods are proposed to estimate the probability density. Predictive distributions are used to assess performance of the model. The methods are illustrated on a real data set.     

采用活性非均匀分布催化剂中变反应器的模拟计算

模拟计算表明:采用活性趋于表层的分布,可以较明显地提高国产中温变换催化剂的内表面利用率,有效地节省催化剂用量,而对“热点”温度没有多大影响。     

Non-steady-state model for kinetics of free radical polymerization, 4. Direct thermal initiation

This paper deals with the non-steady-state kinetics of direct thermal initiated polymerization. The initiation is assumed to be a bimolecular reaction of the monomer. The relationship between the radical concentration and the monomer conversion is rigorously derived. In further treatment a few very close approximations are introduced based on the fact that the number of monomer molecules reacting in the initiation step is much less than that consumed in the propagation step for a process producing high polymer, and the value of the rate constant for propagation or chain transfer is much lower than that for chain termination. Expressions for various molecular parameters, such as molecular weight distribution, number-average and weight-average degrees of polymerization, and dispersity, are given. Several numerical examples are provided.     

用基函数神经网络实现多阈值图象分割

本文介绍了一种用基函数神经网络实现多阈值图象分割的新方法。它从函数逼近的角度研究基于灰度直方图的多阈值分割问题，提出了一种模糊反向传播学习算法，采用该算法的高斯基函数网络能够准确检测直方图中包含的子区域和它们的分布函数，而且速度很快。实验表明本文的方法在实际图象分割中是有效的。     

一种放射CT重建的鲁棒估计算法

放射CT的重建问题可以归结为一种基于泊松分布的最大似然估计（MuxnimumLikelihood,ML），其中最著名的如EM算法[1]，这种算法存在的问题是鲁棒性不强，当观测数据中存在一些出格点（outlier）（如2％）时，EM算法的性能就变得很差。本文分析了算法性能变差的原因，将泊松分布近似为高斯分布后，仿照鲁律估计中的M-估计，构造了一种鲁棒性很强的EM改进算法。实验结果表明，即使在投影数据中混有10％的出格点，仍能重建出满意的结果来。     

镨掺杂分布对1.3μm光纤放大器特性影响

对掺氟化镨玻璃光纤放大器的小信号增益，用广义的高斯近似公式可获得精确的分析表达。文中研究了限制光纤芯层中央部分的镨掺杂对光纤放大器特性的影响，结果发现限制镨掺杂分布能改进光纤放大器的工作效率，且截止波长比芯层全部均匀掺镨的光纤更长。     

Finite Element Analysis of a Moving Boundary Value Problem

A finite element formulation and the solution of a set of nonlinear coupled heat and mass transfer equations for a two-phase system with a moving evaporation interface is presented. The interface condition takes into account the moisture transfer balance at the moving boundary. The finite element results were compared with existing results for a single phase system for model validation. In the two-phase system, the movement of evaporation front has an appreciable effect on the temperature and moisture distribution inside the porous medium during drying. The effect of the nondimensional heat of vapourization parameter γ on the evaporation front, temperature and moisture distribution in porous medium was studied. The higher the value of γ, the slower is the movement of the evaporation front. The temperature decreased and the moisture content increased as the nondimensional vapourization parameter γ increased. This model has potential applications in studying the heat and mass transfer characteristics in food and biomaterials.     

大型储罐的罐壁设计

根据十万立方米浮顶油罐罐壁板应力测试的实测结果，运用应力分析程序，对罐壁板应力分布情况进行了分析和比较，通过调整罐壁板的厚度，使新设计储罐的罐壁板应力分布更加均匀、合理。     

等距码的对偶距离分布及其性质

本文主要讨论了等距码的对偶距离分布及其性质，然后利用这些性质将［１］中的某些结果推广到ｑ元等距码情形，并得到了其对偶距离分布的递推关系式，最后，本文给出了ｑ元等距码的码字数目的一个上界。     

分子结构对增塑聚氯乙烯性能的影响

研究了聚合度、分子量分布和支化结构对增塑聚氯乙烯加工流变性能和物理力学性能的影响。结果表明，增塑ＰＶＣ的加工流变性能随聚合度的增加而恶化；拓宽分子量分布和引入支化结构均有利于加工流变性能的提高；增塑ＰＶＣ的拉伸强度随聚合度的增加而提高，而压缩永久变形却随之减小；分子量分布对物理力学性能的影响不大；支化ＰＶＣ的拉伸强度略有下降。