中国突发性地质灾害分布的统计规律

本文根据地域间突发性地质灾害发生的相似性,将全国区划为9个自然区域;然后依据统计学的大数原理,建立了概率模型分析地质灾害在9个自然区域的分布特点以及隶属于各自然区域的所有省、市和自治区的地质灾害分布特征。结果显示不同类型与不同的区域之间存在着不同程度的强相依性,分布格局明显。     

基于MGPD模型的地质灾害风险的统计度量

利用扩展BurrⅫ分布构建了改进的广义帕累托分布模型—MGPD模型(Meliorated Generalized Pareto Distribution),由此得到了地质灾害损失的在险风险值和最大可能损失估计值。以湖南省娄底市地质灾害损失数据实证分析,结果显示:MGPD模型在刻画地质灾害损失数据时,比GPD模型的精度更高,具有更广泛的适用性。     

M/T-SPH/1排队平稳指标分布的尾部衰减特征

讨论M/T-SPH/1排队平稳队长分布和平稳逗留时间分布的尾部衰减特征,其中T-SPH表示可数状态吸收生灭过程吸收时间的分布。在分布PGF和LST的基础上,给出了两个平稳分布衰减规律的完整分析.结果表明,当参数取不同值时,平稳队长与平稳逗留时间的尾部具有三种不同类型的衰减特征.     

巴丹吉林沙漠东缘不同粒级砂土空间分布的多重分形研究

为了探讨不同粒级砂土(细砂,中砂,粗砂)空间变化的特点,引入多重分形理论,揭示粒级砂土随着土层深度空间变化情况.针对土壤属性做了空间变异性分析,利用Pearson相关系数考察了不同深度下粒级砂土分布密度之间的关系,进一步做了多重分形特征判断,表明不同粒级砂土空间分布存在多重分形的变化特征.又利用广义分形谱探讨了相应的多重分形特点.多重分形谱表明,不同土层深度下,粒级砂土空间变异性增强,同时密度空间分布的变异性依赖高密度分布区的程度增强.     

基于MCMC模拟的MGPD模型及其在地质灾害风险度量中的应用

基于马尔科夫链蒙特卡洛(简记为MCMC)模拟的参数贝叶斯估计,对改进的广义帕累托分布(简记为MGPD)模型进行了优化,并利用该模型得到了地质灾害损失的在险损失值(简记为VaR)和条件损失值(简记为CVaR).以湖南娄底市地质灾害损失数据进行实证分析及模型适应性检验,结果表明:优化后的模型不仅具有很好的极值数据描述能力,而且具有较强的适用性.     

矩形巷道围岩应力分布特征

为得出矩形巷道围岩应力分布特征,根据复变函数理论推导得出了矩形巷道围岩应力解析式并采用巷道围岩沿线分布和matlab软件对巷道围岩应力进行仿真分析两种方法对巷道围岩应力分布特征进行分析.以沿水平线φ=0的围岩应力分布规律为例进行巷道围岩沿线应力分布分析,得出了侧压系数对巷道径向应力和环向应力大小影响规律;侧压系数对径向应力和环向应力峰值出现位置影响规律;围岩至巷道右帮距离与径向应力和环向应力变化规律.以侧压系数对围岩应力空间分布影响为例采用matlab软件对巷道围岩应力仿真分析,得得到了巷道围岩应力空间分布图,可比较方便直观的得到巷道围岩任意位置应力分布情况,还得出了不同巷道尺寸及不同侧压下应力集中系数峰值位置均出现在矩形巷道的四角上.研究结果为巷道支护设计提供了理论依据.     

对M/T-SPH/1排队平稳指标的进一步研究—数值计算与渐近分析

讨论M/T-SPH/1排队平稳队长分布的数值计算,以及平稳队长和逗留时间分布各阶矩的数值计算及渐近分析.其中T-SPH表示可数状态吸收生灭链吸收时间的分布.在分布PGF和LST的基础上,首先给出了计算平稳队长分布,平稳队长以及逗留时间分布各阶矩的数值结果的递推公式.其次还讨论了平稳队长及平稳逗留时间分布各阶矩的尾部渐近特征.结果表明当参数取不同值时,两个指标尾部具有三种不同类型的衰减方式.最后还用数值例子检验了方法的有效性.     

植物种子风传播核的概率模型

种子风传播过程是植被地理分布及其对环境变化响应的关键.植物种子风传播核是种子风传播能力的重要特征.建立种子传播核模型对准确预测植被地理分布十分关键.基于风速的对数正态分布和维布尔分布模式,推导出了植物种子风传播核的两种概率模型,并指出了计算步骤.同时,以6种蒲公英为例,分析了不同植物在不同地点和不同月份种子风传播核特征.算例表明,不同植物、不同地点和不同时间种子风传播核的差异,并且种子风传播强度在风速为对数正态分布下比维布尔分布下要高,传播核强度较高最大传播距离在维布尔分布下比对数正态分布下要远.种子风传播模型研究未来需要考虑近地面层空气动力学参数和风速分布其它模式.     

基于风速随机变量的植物种子风传播距离概率模型

种子风传播过程是植被地理分布及其对环境变化响应的关键.植物种子风传播距离是种子风传播能力的重要特征.建立种子风传播距离模型对准确预测植被地理分布变化十分关键.首先基于物理学中动力学原理,推导出了种子风传播动力学方程,基于风速对数正态分布和维布尔分布模式,推导出了植物种子风传播距离概率模型,并给出了计算步骤.同时,以6种蒲公英为例,分析了不同植物在不同地点和不同月份种子风传距离的概率密度特征.算例表明,不同植物、不同地点和不同时间种子风传播距离概率密度不同,并且种子风传播距离概率密度在风速为对数正态分布下比维布尔分布下要高,传播距离概率密度较高最大传播距离在维布尔分布下比对数正态分布下要远.未来研究种子风传播距离模型需要考虑近地面层空气动力学参数和风速分布其它模式.     

M-Copula模型在金融时间序列分析中的研究与应用

研究了M-Copula模型的建模方法及应用.运用EM算法估计模型的参数,得到相应的统计结果.并利用M-Copula对上证综指和深证成指做了相关分析.通过分析两样本数据的特征,均建立了GARCH-t的边缘分布模型;根据两个对数收益率序列之间的相关特性,选取M-Copula模型对其相关结构进行建模分析,因M-Copula综合了不同Copula的特点,所以分布形式更加灵活,描述数据的厚尾和相关性特征的能力更突出,效果比单一的Copula更好.     

A Green’s function model for the analysis of laser heating of materials

An analytical model based on Green’s function method is developed to analyze the temperature distribution and heated regions in a material irradiated by a high-energy laser beam. The model is multi-dimensional, transient and incorporates different types of beam characteristics and boundary conditions. The multi-dimensional integration formulas in the Green’s function solution equation are evaluated using an adaptive numerical integration algorithm. A parametric study is conducted to show the effect of various laser beam parameters and material properties on the laser heating process.     

Analyzing right-censored length-biased data with additive hazards model

Length-biased data are often encountered in observational studies, when the survival times are left-truncated and right-censored and the truncation times follow a uniform distribution. In this article, we propose to analyze such data with the additive hazards model, which specifies that the hazard function is the sum of an arbitrary baseline hazard function and a regression function of covariates. We develop estimating equation approaches to estimate the regression parameters. The resultant estimators are shown to be consistent and asymptotically normal. Some simulation studies and a real data example are used to evaluate the finite sample properties of the proposed estimators.     

含有治愈部分的可加危险模型的估计

最近可加危险(AH)模型被广泛地应用于生存分析数据,模型的协变量可以假设为时间独立或时间相关的.基于混合治愈模型,有界累计危险治愈模型和"不正确"的比例危险模型.本文将上述的可乘危险模型延伸到可加的危险模型,这里的模型可以允许含治愈部分的生存数据的存在."不正确"的AH模型的识别和参数估计也将在本文给出讨论.     

Statistical analysis of the generalized linear proportional hazards model

We analyze finite sample properties of modified partial likelihood estimators of parameters for the generalized proportional hazards model by simulation. A goodness-of-fit test for the proportional hazards model against the GPH model is proposed. Lung cancer data are analyzed. Bibliography: 7 titles.Published in Zapiski Nauchnykh Seminarov POMI, Vol. 294, 2002, pp. 5–18     

Overlapping design and build cycles in product development

Product development time is a major component of the total time it takes to bring a product to market. Overlapping is the process of starting a downstream operation (such as manufacturing) before completing the upstream operation (such as design). There is an inherent risk in overlapping as a build-assembly may not be compatible with a subsequent assembly, requiring reconciliation work on the overlapped tasks. The ensuing trade-off between this risk, and the project time saved, is shown to possess some unique characteristics among different modes of overlapping. We study the optimization model of the trade-off in general, and analyze some special cases to establish interesting insights on overlapping. These include: advantage of overlapping in different scenarios, special properties of optimal overlap, and impact of parameter values on overlapping decisions. We show that several properties of a single overlap can be extended to multiple overlaps, and establish a limit to overlapping based on administrative cost of implementing overlap and shapes of cost functions. The critical role of the administrative cost (of coordinating overlaps) is also borne out by the model.     

左截断双删失数据下可加可乘风险率模型的估计

在生存分析中,可加可乘风险率模型常用来研究协变量对初始事件和终止事件之间持续时间的影响效应。在本文中,我们考虑了在初始事件存在部分区间删失,同时终止事件存在左截断右删失的情形下,持续时间的可加可乘风险率模型的估计问题。我们提出了一个两阶段估计过程来估计模型的回归参数。并通过模拟分析验证了估计的大样本性质。最后利用该方法分析了恶性黑色素瘤手术治疗数据。     

Remarks on the Analytic Centers of Convex Sets

In this paper we analyze properties of the analytic centers for perturbed convex problems. In particular we study the continuity of convex regions with respect to such characteristics as boundedness, full dimensionality and the existence of an analytic centre. Furthermore, we state the necessary and sufficient conditions for the existence of the analytic centre for possibly unbounded regions and we show that the analytic centre is identical for any minimal representation of the convex set. We are also concerned with the limiting behaviour of the central path as it approaches the set of optimal solutions of the problem.     

区域承接产业转移的产业空间双网络分析——基于金融发展、资源禀赋因素比较视角

利用社会网络分析法,对金融发展、资源禀赋影响下区域承接产业转移的产业空间双网络进行对比分析,结果显示:金融发展、资源禀赋两因素对产业空间网络均产生了正向作用。同时,在产业空间网络中我国各区域基本都具备承接产业转移发展的条件。进一步,经过QAP检验得出两网络呈正相关关系。最后,通过构建2-模网确定了影响我国区域承接产业转移的核心成员(省市、自治区)以及我国区域承接产业转移的合理路径,反映到政策上即发达区域资源禀赋的合理布局以及欠发达区域金融发展的大力支持是目前构建现代化产业体系的关键,本文的研究结论以期为我国区域承接产业转移提供理论借鉴。     

Parametric resonance of a FG cylindrical thin shell with periodic rotating angular speeds in thermal environment

Parametric resonance of a functionally graded (FG) cylindrical thin shell with periodic rotating angular speeds subjected to thermal environment is studied in this paper. Taking account of the temperature-dependent properties of the shell, the dynamic equations of a rotating FG cylindrical thin shell based upon Love's thin shell theory are built by Hamilton's principle. The multiple scales method is utilized to obtain the instability boundaries of the problem with the consideration of time-varying rotating angular speeds. It is shown that only the combination instability regions exist for a rotating FG cylindrical thin shell. Moreover, some numerical examples are employed to systematically analyze the effects of constant rotating angular speed, material heterogeneity and thermal effects on vibration characteristics, instability regions and critical rotating speeds of the shell. Of great interest in the process is the combined effect of constant rotating angular speed and temperature on instability regions.