随机排列方法检验方差变点

给出了随机排列的主要性质及证明。构造了随机排列检验方差变点的统计量。以GARCH(1,1)过程为例,模拟比较了随机排列方法与近似极限分布方法关于方差变点检验的临界值。应用随机排列方法检测人民币兑美元汇率的变点,并与惩罚对比函数方法作比较。模拟与实证结果均表明随机排列方法检验方差变点是灵活有效的。     

基于CVaR的债券投资组合模型

针对债券投资组合中的风险度量难题,用CVaR作为风险度量方法,构建了基于CVaR的债券投资组合优化模型.采用历史模拟算法处理模型中的随机收益率向量,将随机优化模型转化为确定性优化模型,并且证明了算法的收敛性.通过线性化技术处理CVaR中的非光滑函数,将该模型转化为一般的线性规划模型.结合10只债券的组合投资实例,验证了模型与算法的有效性.     

采用蒙特卡洛方法寻找匀质平板的质心

本文介绍基础物理教学中引入蒙特卡洛方法的一个实例 .教学实践表明 :既可使学生学习到物理知识 ,又可开阔他们的知识面 ,有利于学生科学素质的培养     

Monte Carlo Simulation on Coordinated Movement of Kinesin and Dynein Motors

Kinesin and dynein are two important classes of molecular motors which are responsible for active organelle trafficking and cell division. They can work together to carry a cargo, moving along the microtubule in a coordinated way. We use Monte Carlo method to simulate the dynamics of this coordinated movement. Based on four essential assumptions, our simulations reproduce some features of the recent in vivo experiments. The fast moving speed of the cargo is simulated and the speed distribution is presented.     

非序列光线追迹程序照度分布计算的随机误差分析

以点光源在平面上产生的照度分布为基准,研究了非序列光线追迹程序计算照度分布时的随机误差.结果表明,随机误差和计算面上网格的划分方式无关,和各网格内光线条数的平方根成反比.给出了网格内光线条数与随机误差之间的关系公式.根据该公式,为了使结果的标准差优于2%,则单个网格内的光线条数需要大于1 000条.根据具体计算目标,合理选择网格划分方法及追迹光线条数才能获得合理的计算结果.     

Monte Carlo方法在高分子科学中的应用

介绍了Monte Carlo方法的历史及其特点,并描述了它在现代高分子科学研究中的广泛应用情况,并对其前景作了一些展望。     

槽式太阳能聚集阵的间歇跟踪聚光特性

对日间歇跟踪的太阳能聚光电池阵是一种技术难度与效益平衡性好的航天器太阳能电源技术,研究该类空间太阳能聚集系统的在轨聚光特性及间歇跟踪控制方法具有重要意义.本文基于卫星轨道特性与槽式反射聚集阵的光传播特性分析,建立了太阳能聚集阵对日间歇跟踪的时间步长计算方法,考虑太阳光线不平行度等因素影响.以某太阳同步圆轨道卫星上的反射聚集阵为例,采用蒙特卡洛法模拟其间歇跟踪下的聚光特性.结果表明,采用所提出的时间步长计算方法进行聚集阵的对日间歇跟踪,可获得满意的聚光效果.     

Supersolid Phase in One-Dimensional Hard-Core Boson Hubbard Model with a Superlattice Potential

The ground state of the one-dimensional hard-core boson Hubbard model with a superlattice potential is studied by quantum Monte Carlo methods. We demonstrate that besides the CDW phase and the Mott insulator phase, the supersolid phase emerges due to the presence of the superlattice potential, which reflects the competition with the hopping term. We also study the densities of sublattices and have a clear idea about the distribution of the bosons on the lattice.     

Track reconstruction in the BESⅢ muon counter

The reconstruction algorithm for BESⅢ Muon Counter, MucRecAlg, is developed with the objectoriented language C＋＋ in BESⅢ offine software environment. MucRecAlg consists of the following functions：to find track seeds either from extrapolation of tracks in the main drift chamber or from the fired strips in muon counter, to select fired strips associated to the candidate tracks, to fit the candidate tracks with a linear or quadratic function and to calculate other parameters of the tracks for muon identification. Monte Carlo samples are generated to check the performance of the reconstruction package, such as reconstruction effciency, muon remaining rate and pion rejection rate, etc. The preliminary results show that the pion rejection rate is around 3%—4% while the muon remaining rate is better than 90% in 0.4—1.6 GeV/c momentum region, which meets the requirement as shown in the design report.     

Evaluation of particle acceptance for space particle telescope

The particle acceptance instead of the G-factors has been introduced for a particle telescope. The particle acceptance of a telescope module TEST is simulated by using the GEANT4 Monte-Carlo package. The results are presented and explained.