基于MCMC的分位AR模型的贝叶斯单位根检验研究

针对频率统计方法存在不连续的置信区间以及在小样本情况下检验势比较低的问题。把非对称Laplace分布表示成正态分布和指数分布的线性组合,推导了不同先验分布情况下参数的最大后验密度置信区间,并构造了分位回归单位根检验的贝叶斯因子,实现了对非平稳时间序列的局部单位根检验。仿真分析表明贝叶斯分位回归方法是一种稳健全面的单位根检验方法。对我国居民消费价格指数的实证研究发现,我国居民消费价格指数表现出局部的持续性,在分布的下尾部不受普通冲击的影响,但在分布的上尾部受普通冲击的影响。     

支持向量机和神经网络在粗糙面参数反演中的比较

首先介绍支持向量机和神经网络方法及其在内部网络训练上的不同.分别利用支持向量机和神经网络对高斯粗糙面的均方根高度和相关长度进行反演.通过仿真结果和误差对比分析,发现在小样本情况下,支持向量机的反演结果比神经网络好,而在具有大量样本的情况下,神经网络的反演精度有显著提高,而且反演时间比支持向量机少很多.     

特殊矩阵的特殊对角化方法研究

主要讨论了矩阵只有两个特征根时的对角化问题,方法简捷,效果耐人寻味.     

A combined model for pseudorapidity distributions in p-p collisions at center-of-mass energies from 23.6 to 7000 GeV

In p-p collisions, the charged particles produced consist of two leading particles and those frozen out from the hot and dense matter created in the collisions. The two leading particles are in the projectile and target fragmentation regions, respectively, which, in this paper, are conventionally supposed to have Gaussian rapidity distributions. The hot and dense matter is assumed to expand according to unified hydrodynamics, a hydrodynamic model which unifies the features of the Landau and Hwa-Bjorken models, and freeze out into charged particles from a space-like hypersurface with a fixed proper time of τFO. The rapidity distribution of these charged particles can be derived analytically. The combined contribution from both leading particles and unified hydrodynamics is then compared against experimental data from a now available center-of-mass energy region from 23.6 to 7000 GeV. The model predictions are consistent with experimental measurements.     

鸟体姿态对结构抗鸟撞性能的影响

低于现行标准规定能量的大量鸟撞事故中,航空结构仍然发生实质性破坏的情况,说明只考虑鸟体的质量和速度不足以保证飞机安全。本文中针对弹性平板、雷达罩及机翼前缘等飞机典型结构,开展了不同姿态鸟体的鸟撞分析研究。分析结果发现,鸟体姿态对结构的抗鸟撞性能有比较显著的影响,不同的结构特点反映的响应规律也不同:对吸能结构,姿态角越大,吸收的能量越多,被保护的结构就越安全;而对承力结构,姿态角越大,高应力区域越大,结构就越危险。因此,在结构的抗鸟撞安全性评估中,除了完成特定姿态鸟体的鸟撞实验,针对危险工况还应通过数值分析评估不同鸟体姿态的结构撞击响应,进一步确保结构的抗鸟撞能力。     

Preparation and Characterisation of Thermoplastic Starches from Cassava Starch,Cassava Root and Cassava Bagasse

Summary: Thermoplastic starches (TPS) based on cassava starch have been produced by extrusion at 120 °C, using glycerol as plasticizer. Three forms of cassava starch were employed, viz: cassava root (CR), cassava bagasse (CB) and purified cassava starch (PCS). The main differences between these are the presence of sugars and a few fibres in CR and high fibre concentration in CB. Conditions of processing and characteristics such as amylose and fibre content, crystallinity, water absorption and mechanical behaviour in the tension x deformation test were evaluated. The results demonstrated that the PCS and CR had amylose contents consistent with literature values (14–18%) and that CB is a material constituted mainly by amylopectin. It was found that fibres in high proportions (as in the bagasse) can confer reinforcement properties and are thus able to generate natural composites of TPS with cellulose fibre. The sugars naturally found in the root reduce the elongation of the TPS under tension. The PCS and CR TPS were stable with respect to indices of crystallinity after processing; and during a period of 90 d in a relative humidity of 53%, while the CB TPS tended to vary its crystallinity, probably because its amylose chain had low degree of polymerization.     

Flocks and Partial Flocks of Hyperbolic Quadrics via Root Systems

We construct three infinite families of partial flocks of sizes 12, 24 and 60 of the hyperbolic quadric of PG(3, q), for q congruent to -1 modulo 12, 24, 60 respectively, from the root systems of type D ₄, F ₄, H ₄, respectively. The smallest member of each of these families is an exceptional flock. We then characterise these partial flocks in terms of the rectangle condition of Benz and by not being subflocks of linear flocks or of Thas flocks. We also give an alternative characterisation in terms of admitting a regular group fixing all the lines of one of the reguli of the hyperbolic quadric.     

Modified Lagrangian and Least Root Approaches for General Nonlinear Optimization Problems

Abstract In this paper we study nonlinear Lagrangian methods for optimization problems with side constraints.Nonlinear Lagrangian dual problems are introduced and their relations with the original problem are established.Moreover,a least root approach is investigated for these optimization problems.     

A Weighted Enumeration of Maximal Chains in the Bruhat Order

Given a finite Weyl group W with root system , assign the weight to each covering pair in the Bruhat order related by the reflection corresponding to . Extending this multiplicatively to chains, we prove that the sum of the weights of all maximal chains in the Bruhat order has an explicit product formula, and prove a similar result for a weighted sum over maximal chains in the Bruhat ordering of any parabolic quotient of W. Several variations and open problems are discussed.     

Stable iterations for the matrix square root

Any matrix with no nonpositive real eigenvalues has a unique square root for which every eigenvalue lies in the open right half-plane. A link between the matrix sign function and this square root is exploited to derive both old and new iterations for the square root from iterations for the sign function. One new iteration is a quadratically convergent Schulz iteration based entirely on matrix multiplication; it converges only locally, but can be used to compute the square root of any nonsingular M-matrix. A new Padé iteration well suited to parallel implementation is also derived and its properties explained. Iterative methods for the matrix square root are notorious for suffering from numerical instability. It is shown that apparently innocuous algorithmic modifications to the Padé iteration can lead to instability, and a perturbation analysis is given to provide some explanation. Numerical experiments are included and advice is offered on the choice of iterative method for computing the matrix square root.